Punting - why value is the key

Colonel Sanders

There was a discussion on the Grandouet thread about the fundamental aspect of betting, that was off topic and I think a few posters get annoyed when threads are derailed into side shows. Therefore I decided to start a thread of it's own. First up I'll have a stab at explaining why you cannot make money in the long run unless you bet at prices that under estimate the chances of your selection winning.

*apologises if I go a bit abstract and/or into a bit too much detail. Feel free to question anything I write

First a bit of house keeping

I will use the shorthand P(X) to represent the probability of event X happening eg if I toss a coin P(H) = the probability the coin lands on heads

For any numerical examples I will use the example of tossing a "fair" coin. By "fair" I mean P(H) = P(T) = 50%. For the more pedantic we can either assume that the probability of it landing on it's side is zero, or if you like, given that it lands on it's side all bets are void with stakes returned

For illustration of some of the points I will assume I have found a bookmaker that is betting on the coin toss but has incorrectly priced his market. I assume the following;

1. He bets 4/6 Tails.
2. He bets 11/10 Heads

Lastly I assume that each coin toss is independent of the last, i.e. knowing the result of the last coin toss gives me no further information about the outcome of the next coin toss

Concept 1: Expected Value

The expected value (EV) represents the weighted average of all possible values that a variable can take

If we take the coin toss market our friendly bookmaker has priced. If I place a €1 bet on heads at 11/10 my cashflow is as follows;

1. I will win €1.10 with probability 50%
2. I will lose €1 with probability 50%

From our definition of expected value we see this bet has positive expected value, or +EV. I will call this a "good" bet

+1.1*0.5 - 1*0.5 = 0.05

This means that if I place a large number of bets at 11/10 I will can expect to win 50% of my bets and lose the other 50%. However the profit on the winning bets more than cancel the losses on the the losing bets. So I play this market over and over again I should come out in front to the tune of 5% of my turnover.

If I place a €1 bet on tails at 4/6 my cashflow is as follows;

1. I will win €0.67 with probability 50%
2. I will lose €1 with probability 50%

From our definition of expected value we see this bet has negative expected value, or -EV. I will call this a "bad" bet

+0.67*0.5 - 1*0.5 = -0.1667

This means that if I place a large number of bets at 4/6 I can still expect to win 50% of my bets and lose the other 50%. However the profit on the winning bets is not enough cancel the losses on the the losing bets. So I play this market over and over again I should lose, to tune of 16.67% of my turnover.

So we should only make "good" bets

Concept 2: Variance

So we should only place "good" bets. However can we still lose money? Unfortunately the answer is Yes. Shrewd punters can still lose money over a finite series of bets and poor gamblers can can still win money over a finite series of bets due to variance

If we place only 1 bet we will lose money if tails comes up and will win if heads comes up. So our actual P&L will be either +€1.10 or -€1. Neither is equal to the EV mentioned above because of statistical variation or variance. So we win or lose money with probability 50%. Doesn't sound as good now does it?

However lets assume we place 10 bets on this market. We will win money if 5 or more heads come up. We will lose money if 4 or less heads come up. These probabilities can be evaluated using what is called a binomial distribution. I won't go into this but excel has a function to calculate these probabilities and there are numerous online calculators

If we place 10 bets we will win money with probability 62.3%. Better but not all that great you may think

If we place 100 bets we win if 48 or more coin tosses come up heads. Therefore the probability we win money (again from a binomial distribution) is 69.1%

If we place 100 bets we win if 477 or more coin tosses come up heads. Therefore the probability we win money (again from a binomial distribution) is 93.1%

What is happening is that the more "good" bets we place the higher the probability we win money overall. And also the more good bets we place the closer our profit on turnover gets to our EV

This leads us onto the fundamental mathematical theorem underlying this explanation.

Concept 3: The Law Of Large Numbers

The law of large numbers says, in lay mans terms, that the larger the number of "good" bets we place the closer and closer our profit on turnover will get to our EV (5% in the coin toss example)

If we could place an infinite number of bets we would be certain to achieve our EV (as a % of turnover), no more no less. i.e. in the coin toss example if we could place an infinite number of bets at 11/10 the probability we would win 5% on turnover is 1.

A simulated example

I simulated 160,000 coin tosses in excel (using the Rand() function) and calculated the profit on turnover. The attached graph shows that our profit on turnover (the purple/blue line) gets closer and closer to 5% (the grey line) as the number of coin tosses, and bets get larger. In this case we actually have a bad run of luck over the first quarter or so tosses but these losses even out over the long run. If we could simulate an infinite number of coin tosses the law of large numbers ensures that we would achieve a profit of 5% on turnover

FWIW I got 50.03% simulated tails and 49.97% simulated heads

Horse races aren't coin tosses!

That's all well and good, in the situation described above we know explicitly P(H) and P(T) before we place our bets. This is not the case we horse racing. Before a race we can never know the probability a horse will win a race with more than 1 runner with certainty. This is where knowledge of form, going, distances etc come in. We make an implicit estimation of the chances our horse wins, if the bookmaker offers better odds we think we should be getting we bet. There are a few metrics you can use to keep track of your ability top price up the horses you are backing. I keep track of P&L, Return on turnover, price taken v Betfair SP etc etc

I calculated the probabilities above very quickly so may have made a few small numerical error but the underlying theory is sound. Any queries fire away!

*if anyone is interested, for the coin toss example since the set up and probabilities are the same for each coin toss the weak law of large numbers suffices. In the case of betting on sports where each event and probabilities is different we must apply a more complicated form called Kolmogorov's Strong Law of Large Numbers

Colonel Sanders

I tried to post an excel spreadsheet with the results and graph of the above simulations but it exceeds the max size of an xls file

Is there a way I can post a graph???

ste2010

Colonel Sanders wrote: »

I tried to post an excel spreadsheet with the results and graph of the above simulations but it exceeds the max size of an xls file

Is there a way I can post a graph???

Can you zip it or make it public on skydrive and post the link

Colonel Sanders

ste2010 wrote: »

Can you zip it or make it public on skydrive and post the link

I'll have a look at that or the likes of google docs later, I'm watching the Darts!

If worst comes to the worst I'll print screen and copy it into a Word Doc that hopefully won't be too big!!

ft9

Off to the gambling forum with ye

Colonel Sanders

ft9 wrote: »

Off to the gambling forum with ye

I'm hoping the mods will be ok leaving it here

A few threads get derailed talking about value, they could be re-routed to here

Colonel Sanders

Colonel Sanders wrote: »

Shrewd punters can still lose money over a finite series of bets and poor gamblers can can still win money over a finite series of bets due to variance

A good example is the lotto (from a pure gambling point of view), many people become millionaires playing it but only around 50% of the money bet is paid out as prizes.

Paying out 50% is like betting into a 200% over round :eek:

ft9

Colonel Sanders wrote: »

I'm hoping the mods will be ok leaving it here

A few threads get derailed talking about value, they could be re-routed to here

I'm only messing with ya, best of luck with it, just not my thing.

Hulk Hands

Anyone playing Roulette or Slots as a serious way to make money instantly lose my respect because of this. Actually, anyone playing them at all really.

Hate to be nitpicking as there was clearly effort put into that post, but it's complete and utter common sense. The majority of readers here already know all that, while the rest are either too stupid to grasp the concept or too weak of mind to stop punting who they think will win regardless of price. i.e you won't be changing too many people's outlooks

Colonel Sanders

Hulk Hands wrote: »

Hate to be nitpicking as there was clearly effort put into that post, but it's complete and utter common sense.

Agreed, not everyone agrees however

Hulk Hands wrote: »

The majority of readers here already know all that, while the rest are either too stupid to grasp the concept or too weak of mind to stop punting who they think will win regardless of price. i.e you won't be changing too many people's outlooks

Probably not to be fair, but as I said a lot of threads get side tracked with value discussions etc. They can be directed here and leave the other threads for the topic at hand

Hulk Hands

Colonel Sanders wrote: »

Agreed, not everyone agrees however



Probably not to be fair, but as I said a lot of threads get side tracked with value discussions etc. They can be directed here and leave the other threads for the topic at hand

Sorry Colonel, I actually hadn't seen the Grandoeut thread

WOW.......

Nulty

Col you mad bastard

BQQ

Ok, I studied statistics once upon a time, so I have no issue with the maths.

However, just because you think you're getting value doesn't mean you are.
You may just be reading the form badly.

So, if you thought a given horse had a 12.5% chance of winning you would back it at 8/1 and you wouldn't at 6/1, but if your reading of the form was wrong and the horse had a true chance of 5%, you would obviously lose in the long run even though you're only backing what you believe are value bets.

Before the Grandouet thread became the bash Huntley thread, he tried to make the point that value is subjective and therefore a correct analysis of the form is more important than value since your idea of value is derived from your analysis of the form.

Colonel Sanders

BQQ wrote: »

However, just because you think you're getting value doesn't mean you are.

You may just be reading the form badly.

Couldn't agree more, and nowhere have I, or anyone else that I can see, have said otherwise. As I posted there are a few metrics you can track to see if you are getting it right

Powers often have huge priced specials on soccer, I am not the biggest soccer fan. I price then at Betfair lay prices + an additional 10%. If they are bigger than that I back them. I keep track of:

1. The price Powers offered
2. The price implied by Betfair lay prices
3. EV assuming 2 is correct
4. P&L assuming you have their max on and also you bet them all to win €200 (€200 as the lowest price I've ever seen offered is evs, and the max bet is usually €200)

BQQ wrote: »

Before the Grandouet thread became the bash Huntley thread, he tried to make the point that value is subjective and therefore a correct analysis of the form is more important than value since your idea of value is derived from your analysis of the form.

Here's what he said that I responded to:

You can just as easily lose in the long term by selecting overpriced horses.

Not true

Value has no importance if you are consistently picking losers. There is an array of evidence to support that

picking winning selections is the most fundamental aspect of successful gambling, not value as you seem to think.

There is no mathematical fact which states that value is the most fundamental aspect to successful gambling.

He had no evidence forthcoming, it also violates the law of large numbers

No problem with arrogance provided it is backed up.

He hasn't backed up his arrogance on this topic, he has dug a bigger hole

Do you understand the idiocy in stating that value is the most important aspect of successful gambling? The only value that counts is a winning selection.

Colonel Sanders

Response to this post, the Grandouet thread isn't the place

Huntley wrote: »

What is the marker to decide whether the bookie is wrong? The only marker is the selection winning

P&L, RoI, strike rates, actual versus expected results over a long period of time

Huntley wrote: »

It is all subjective, you may think that you are backing horses which are 'value', it may not be to someone else.

I agree and have said so before

Huntley wrote: »

I determine value by the selection that is most likely to win.

you are wrong to do so, in my coin toss example both outcomes are equally as likely to occur. Only one of the potential bets being offered is value. This is the case win or lose. You really seem to have a problem grasping this concept, you can get good value and lose, you can get poor value and win over the short term

Huntley wrote: »

If a horses price collapses from 20's to 2's and loses does that mean the price is value?

Not necessarily

Huntley wrote: »

What value is there in a losing selection? In my opinion, there is none.

Your opinion (on this topic) isn't worth the virtual paper it's written on

Huntley wrote: »

In contrast to that, if the price collapses from 20's to 2's and the horse wins does that mean that the winning price wasn't value?

It may or may not

Huntley wrote: »

any winning selection is a value bet

No it isn't

Huntley wrote: »

My views aren't as outrageous as much as the clueless masses who descended yesterday hoped

Yes they are

Huntley wrote: »

there is plenty of credibility behind my thinking

Yet you have not presented any. Trust me, there is none

Huntley wrote: »

I don't post here to come across self righteous

You do however, intended or not

Huntley wrote: »

I do it to provide different angles and analysis on races

Tbh I think you do this pretty well

Huntley wrote: »

I can be very dismissive of idiotic opinions, which I most certainly am not accustomed to myself.

Read your posts on the Grandouet thread, that will acquaint you somewhat

Huntley wrote: »

it is no surprise that the piss poor analysis in general is synonymous with the external views about this forum.

Some of your comments on the Grandouet thread are some of the worst, ill thought out musings I have ever come across. As is your refusal to admit you're wrong.

Colonel Sanders

Colonel Sanders wrote: »

A simulated example

I simulated 160,000 coin tosses in excel (using the Rand() function) and calculated the profit on turnover. The attached graph shows that our profit on turnover (the purple/blue line) gets closer and closer to 5% (the grey line) as the number of coin tosses, and bets get larger. In this case we actually have a bad run of luck over the first quarter or so tosses but these losses even out over the long run. If we could simulate an infinite number of coin tosses the law of large numbers ensures that we would achieve a profit of 5% on turnover

FWIW I got 50.03% simulated tails and 49.97% simulated heads

I re-did this experiment. I simulated 500,000 coin tosses in excel. Due to the fact there is a maximum xls size you can post to boards and a graph can have a max of 32,000 data points on a graph I had to summarise the results.

The sheet "Simulations", each entry is the cumulative RoI figure calculated at intervals of 500 simulations. The graph shows the simulated RoI (noisy blue line) versus the 5% EV (straight line). While the results move up and down a good bit they approach 5% as the number of simulations increase

Slattsy

I love excel, probably one of the best inventions ever.

A very decent thread and the OP is on the ball. In my opinion the bookies prices on each horse running are typically 15% less than what would be required for a consistent level stakes punter to break even longterm.

If you accept that the rationale that your happy to back a horse regardless of price if you think it will win makes losing money inevitable unless you are typically backing ten or so horse a year, with a sample size that small it is conceptually likely to put a very decent strike rate together and come put on top If you back 10 to 15 horses a day in the bookies regularly the lack of value will sting you which is why there are so many bookies about

I used to back horse that were in my mind value, I had a crude enough way to calculate what acceptable value was but I only went looking at the prices after detailed study of the form so it was very time consuming yet admittedly moderately successful.

People will have varying views on what value is, a price could be considered value if it's not as short as you think the market should have it, even if its still less than the true odds reflection. Others would only consider a price value when the odds available are in their mind greater than what they reckon the true odds of the horse winning are.

Of course one has to accept that trying to accurately reflect the likelyhood of a horse (not to mention a number of horses on one race) performing to past levels or to the trainers expectations on the day is very difficult, hence the reason most prices aren't value, the bookies build in a level of fat to ensure they won't be caught out too often

Unfortunately, to me it seems much easier to make money laying horse

Not as much fun and the buzz of a winner eclipses the laying experience, I still back quite a few horses for "fun", despite them not being value most of the time

I've only just read beyond the op now, wouldn't have bothered posting tbh if I had read on initially. Very sour tone to it.

Colonel Sanders

Colonel Sanders wrote: »

.................







He had no evidence forthcoming, it also violates the law of large numbers ............

You make repeated reference to the law of large numbers.....

http://www.boards.ie/vbulletin/showpost.php?p=82648984&postcount=198

RoverJames wrote: »

Define what a large number is CS.
If you actually know what you are talking about you have to concede that a selective punter over 50 years may well come no where near backing enough horses for the total to be considered a large number.

It's quite clear to me that a selective punter who interprets form etc in a manner that results in a high strike rate can punt away happily for his lifetime not at all being bothered about the law of large numbers

kiers47

RoverJames wrote: »

You make repeated reference to the law of large numbers.....

http://www.boards.ie/vbulletin/showpost.php?p=82648984&postcount=198




It's quite clear to me that a selective punter who interprets form etc in a manner that results in a high strike rate can punt away happily for his lifetime not at all being bothered about the law of large numbers

While I understand what you are saying. What do you consider a respectable strike rate? It surely depends on average prices.

Lets say for instance
punter A has a 80% strike rate but he only bets on 1/10 shots or lower.
Punter B has a 10% strike rate but only bets on 16/1 + shots.

Who is the better punter? It's fairly easy to see!!

meriwether

Colonel, I remember reading about this theory before in Dave Nevisons book.

His approach is to price the race independently, and he identifies value as being in place if a bookmekrs price is 15% or more larger than his own opinion of the horses price.

However, and this throws me a bit, he bets on every horse which meets this criterion. He bets on horses he logically doesn't think will win (he prices them the rag 66/1 and Betfar offers 100/1).

I find that bizarre.

kiers47

kiers47 wrote: »

While I understand what you are saying. What do you consider a respectable strike rate? It surely depends on average prices.

Lets say for instance
punter A has a 80% strike rate but he only bets on 1/10 shots or lower.
Punter B has a 10% strike rate but only bets on 16/1 + shots.

Who is the better punter? It's fairly easy to see!!

My point is that the law of ilarge numbers doesn't apply
Instead of decent strikerate I should have said decent ROI, I was referring to a selective punter who would be in profit.

meriwether

meriwether wrote: »

Colonel, I remember reading about this theory before in Dave Nevisons book.

His approach is to price the race independently, and he identifies value as being in place if a bookmekrs price is 15% or more larger than his own opinion of the horses price.

However, and this throws me a bit, he bets on every horse which meets this criterion. He bets on horses he logically doesn't think will win (he prices them the rag 66/1 and Betfar offers 100/1).

I find that bizarre.

I often lay horses I think will win if the price is imo too short.

Colonel Sanders

RoverJames wrote: »

Define what a large number is CS.

In most situations you can't definitively say what a large number is. If you make some assumptions on the type of bets you play, how much edge you have you can derive the minimum number of bets necessary such that the probability you win is a certain percentage. I the coin toss example you could use the normal approximation of the binomial distribution

The law of large numbers does not guarantee profit, it formally proves that the more "good" bets you place the more likely you are to get a profit on turnover. The more "bad" bets you place the more likely you are to lose. In the case you could place an infinite number of bets it states a "good" punter will win money with probability 1, a "bad" punter will lose money with probability 1

RoverJames wrote: »

If you actually know what you are talking about you have to concede that a selective punter over 50 years may well come no where near backing enough horses for the total to be considered a large number.

I stated in the OP that a "good" punter can lose money and a "bad" punter can win money (the lotto is an excellent example, some people do become millionaires but playing is an irrational betting strategy)

What I sought to show in the OP was that the key to gambling was value (i.e. betting at prices that under estimate your selections chance) and not merely identifying the most likely winner

Colonel Sanders

Colonel Sanders wrote: »

In most situations you can't definitively say what a large number is.............

An approximation would do

For example, imo, 15 bets a year over 50 years would be 750 bets, not a large number.

4000 bets a year over 5 years, 20,000 bets, a large number

As pointed out value is really quite subjective, more often than not when most self proclaimed good punters reckon they're getting value they're quite likely not Anyone can claim their bets are value, only a few can consistently pick enough winners to maintain a positive ROI

Personally as I've mentioned a few times I lay quite a lot of horses, I back comparatively few, I understand the concept of betting value perfectly.

Your application of mathematical theory to this subject seems to be quite case non specific as you seem to make your argument on the rules of large numbers which may or may not be at all relevant.

Calling a spade a spade you only started the thread to undermine another posters views and opinions

Slattsy

You do like a smiley face RJ.

Slattsy

Slattsy wrote: »

You do like a smiley face RJ.

I'm not meaning to get controvercial or stir sh1t you see so I reckon it's better to lash in a few of them lads to keep the tone cordial, the ole typed words on a screen can be easily misinterpreted.

From what I've gathered since first posting in this thread Huntley made comments along the lines of he isn't bothered by value in bets, the law of large numbers obviously disagrees with this but the example given here isn't overly apt due to the odds reflecting the actual probability, betting isn't like that and no one knows the real probability, imo it can't be quantified but we do know the bookies will try and take 10 to 15% (ish) from the "fair" odds

Imo a selective punter who consistently picks enough winners can bet away happily ignoring value etc as his ROI (buoyed by the edge he has, ie picking more winners than most relative to losers) is sufficient to absorb the bookies margin.

A few can do this, a couple have successfully done so in the form of logs on boards.ie, hard to ignore evidence really imo.


I can see how the selective punter can ignore the "value" concept as they've bought into the bookies getting their margin.

I can't actually fathom how someone can apply the law of large numbers to punting without conceding it may well be non applicable in some cases.

All anyone has to do it take 1000s of SPs (and results of course) within a range and if the sample size is large enough you can see that backing them all for level stakes will result in a loss of 10 to 15% ish of your outlay, take smaller sample sizes and there'll be variation in the results.

Repeat for other SP ranges and the same will show, the shorter the SP the smaller the sample size required.

Colonel Sanders

RoverJames wrote: »

Your application of mathematical theory to this subject seems to be quite case non specific as you seem to make your argument on the rules of large numbers which may or may not be at all relevant.

Calling a spade a spade you only started the thread to undermine another posters views and opinions

If you give a specific example I'll try derive an estimate of this "large number". It would be no more than an approximation however

I fail to see how it isn't relevant. I've posted several times here on how it is. Let me repeat for the last time, an application of the law of large numbers to gambling is you can't make money in the long run if you fail to obtain incorrect prices. As a corollary the probability you win money in either the long run or over a finite period increases the more "good" bets you place. Winning money when you place "bad" bets is pure chance as is losing money when you place "good" bets.

I also think I've posted that value is subjective at least a few times. You can't know the prior probability of a horse race/any other sporting event with certainty. However there are metrics you can use as an indication.

I made this thread to stop threads getting side tracked about value. The Grandouet thread wasn't the only one. In that specific thread my opinion on the subject was called "idiocy". I made an attempt to outline how there was nothing idiotic about it.

Despite your last comment You are spuriously attempting to undermine my opinion and you're not doing a very good job tbh

Colonel Sanders

By the way please give a case where the law of large numbers is not applicable?

Slattsy

I hear you barking big dog.
Value is key clearly, but everyone's opinion of value is different. Because of opinion and assumptions.

I never get too caught up in the statistics side of things. I follow form. If I think a horse should be evens and the bookies price it up at 2/1 then I'll steam in.
The flip side to that is.... If I price a nag up at 3/1 and think it will win, but the bookies price it up at 7/4, do I still back it? If I think it will win then of course I'll back it. That's how I roll.

There are instances where I believe a horse are underpriced and i won't play though. And then I'll look elsewhere or call it a no bet race.

Everyone prices, picks, studies horses in their own unique way.

There was a lot of nonsense posted but as I say, each to their own. I'm like Luxembourg, I try to stay neutral :-)

Pinesky

I agree value is a key component to overall betting strategy but it is not possible to truly compute absolute value before the race ends because there are unknown chance factors , horse is sick , left in stalls , poor ride, fall, brought down, baulkef, trapped on rail , not get clear run etc.Maybe all these variables can be deemed to result in good luck or bad luck.
What I do subscribe to is if you back a horse at 2/1 that should be 1/1 you have a better chance of making money than if you do the reverse but it is not absolutely guaranteed .

Colonel Sanders

Colonel Sanders wrote: »

If you give a specific example I'll try derive an estimate of this "large number". It would be no more than an approximation however
......................................

Despite your last comment You are spuriously attempting to undermine my opinion and you're not doing a very good job tbh

Backing 10 to 15 horses a year over 50 years, total number of bets 500 to 750, large number ?? Yes or No.

I'm not undermining your opinion, I'll agree to be disagreeing with your blanket application of the large number theory to punting.

You can claim I'm not making a good job of it all you like, there are countless statistically valid sampling plans that illustrate the importance of having an appropriate sample size. When looking at attribute data (win or lose for example), the sample size needs to be large to make any conclusions. With variable data and appropriate analysis smaller sample sizes are ideal but the able analyst will appreciate there will be variation is results across pools of data so only a confidence statement will be made.

Colonel Sanders wrote: »

By the way please give a case where the law of large numbers is not applicable?

Anywhere there isn't a large sample size.

As per above 10 to 15 bets a year on UK/Irish horse racing when there are over 10,000 races per year

droidman123

Slattsy wrote: »

I hear you barking big dog.
Value is key clearly, but everyone's opinion of value is different. Because of opinion and assumptions.

I never get too caught up in the statistics side of things. I follow form. If I think a horse should be evens and the bookies price it up at 2/1 then I'll steam in.
The flip side to that is.... If I price a nag up at 3/1 and think it will win, but the bookies price it up at 7/4, do I still back it? If I think it will win then of course I'll back it. That's how I roll.

There are instances where I believe a horse are underpriced and i won't play though. And then I'll look elsewhere or call it a no bet race.

Everyone prices, picks, studies horses in their own unique way.

There was a lot of nonsense posted but as I say, each to their own. I'm like Luxembourg, I try to stay neutral :-)

as usual slattsey, you probably talk the most sense on here. i think this all started with me stating that if i thought a horse that i really fancied in a race was,say,6/1 and i thought he should be 8/1 i would back it anyway.but as usual self styled guru,s on here start making a mountain out of a mole hill. does anyone really think that a load of stats about betting mathemethics would make me change my way of punting?:rolleyes: i have my way and others have theirs.its really is that simple.

premierstone

Here's a link I came accross that might help those who are still adamant that Value is not the single most important factor to sucessfull punting - http://www.geegeez.co.uk/25524/

BQQ

premierstone wrote: »

Here's a link I came accross that might help those who are still adamant that Value is not the single most important factor to sucessfull punting - http://www.geegeez.co.uk/25524/

Value betting is based on making your own tissue prices.
You do that by reading the form.
Do it wrong and your prices are worthless.

Seems to me that reading the form properly is the single most important factor to successful punting.

Something that the writer of that article seems unable to do btw.
He seems to be looking for something to confirm his pre-race analysis, rather than accepting he got it wrong.
He says his selection was the only one making ground in the closing stages off a slow pace, but in fact, the horse than came second was held up and stayed on too. And obviously better than his selection. Maybe he didn't notice as it was one of the horses he categorically ruled out beforehand.

premierstone

premierstone wrote: »

Here's a link I came accross that might help those who are still adamant that Value is not the single most important factor to sucessfull punting - http://www.geegeez.co.uk/25524/

I'm a firm believer in value and tissue prices etc, I aim to lay about 3000 to 4000 short priced horses in 2013 and make a decent wad of cash in doing so. We all know if you lay 3000 to 4000 horses on the exchanges willy nillly you'll lose your arse.

I can completely see how someone can just accept that the bookies are aiming to take 10 to 15% from bets and punt away selectively ignoring the value concept and just back horses (10, 20 ish a year) that they think have an awfully decent chance of winning.

Value is immaterial to them, the result is all they care about.

I'm different, I don't really care about the result, all I'm bothered about is laying at a price I consider to be poor value for the backer. If I do that 200/300 times a month I'm happy, I know 1 in 3 / 1 in 4 of them will win.

If I was only laying 20 horses a year the result would be far more important than the value side of things.

Colonel Sanders

RoverJames wrote: »

Backing 10 to 15 horses a year over 50 years, total number of bets 500 to 750, large number ?? Yes or No.

The law of large numbers does not answer this question, indeed it was never intended to do so.

If by "large number" you mean "guaranteed to win" the number is infinity. The best you can do is say I'm happy to accept a probability of p% that I'll win money and try derive how many bets you need to place

In the example in the OP if you have 750 bets you win with probability 89.9%

If you only punt 4/6 shots that should be 1/2 then if you place 750 bets the probability you win is 99.9943%

Are these probabilities acceptable to your risk tolerance? If so then 750 is "large" if not then 750 is not

RoverJames wrote: »

I'm not undermining your opinion, I'll agree to be disagreeing with your blanket application of the large number theory to punting.

My blanket application was value is key. The law of large numbers merely supports this

RoverJames wrote: »

You can claim I'm not making a good job of it all you like, there are countless statistically valid sampling plans that illustrate the importance of having an appropriate sample size. When looking at attribute data (win or lose for example), the sample size needs to be large to make any conclusions. With variable data and appropriate analysis smaller sample sizes are ideal but the able analyst will appreciate there will be variation is results across pools of data so only a confidence statement will be made.

......

Anywhere there isn't a large sample size.

As per above 10 to 15 bets a year on UK/Irish horse racing when there are over 10,000 races per year

The lack of a large sample size doesn't invalidate the statement that "value is key". Nor does it make the law of large numbers inapplicable

All a small sample size does is:

1. Make the probability a "bad" punter can manage to win larger and conversely the probability a "good" punter can lose larger
2. Make assessing whether you are correctly sizing up a bookmakers odds accurately more difficult

All the law of large numbers says is that the larger number of bets the closer your RoI will be to EV (whether that EV is positive or negative)

Even if you only have 1 bet every 5 years why would you not place that bet at favourable odds?

I bet there are people who have won the lotto after playing a handful of time, yet playing the lotto is as irrational a gambling strategy as you can get

Colonel Sanders

Colonel Sanders wrote: »

........



My blanket application was value is key. The law of large numbers merely supports this
.............

Again I'll repeat, for someone selectively backing 15 horse per annum, value isn't key, the result is. They just accept the bookie have taken their cut (many would expect all of their selections to be priced consistently "unfairly") and the punter ploughs on. They presume the price is no better or no worse than most of the other days prices and they feel they have an edge due to their knowledge or interpretation of the race and the other horses in it.

It's very easy to speel on about value, in punting circles it's the done thing. It's not so easy to show positive ROI over a period, a few on here have done so, most haven't.

All this talk of value and favourable odds is merely bluster unless you can demonstrate you have the capability to find value and favourable odds. It's like people talking about curling free kicks into the corner around the wall.

There are imo as few people backing selections that are value consistently as there are people with +ive ROIs, there are plenty people who talk about both though, hurler on the ditch for want of a better analogy

You mention .....

Colonel Sanders wrote: »

........

All a small sample size does is:

1. Make the probability a "bad" punter can manage to win larger and conversely the probability a "good" punter can lose larger
2. Make assessing whether you are correctly sizing up a bookmakers odds accurately more difficult.... ...........

Can you not concede there are other options, such as ...

3. The punters small number of bets (15) over the period showed they might well actually know more about racing than the majority of punters due to a high percentage of them winning.

Or is it a case of it's just option 1 or option 2?

I'm becoming increasingly bewildered at your reluctance to admit certain posters might be better at picking winners than the rest of us.

You also haven't offered an approximation of what a large number would be, I asked would you consider 15 bets a year over 50 years a large number, the answer to that is Yes or No.

Huntley

A final few words on the issue here as very few people seem to be acknowledging my opinion constructively.

BQQ wrote: »

Value betting is based on making your own tissue prices.
You do that by reading the form.
Do it wrong and your prices are worthless.

Seems to me that reading the form properly is the single most important factor to successful punting.

Precisely. Really enjoyed your take on the issue here aswell http://www.boards.ie/vbulletin/showthread.php?t=2056832228&page=12

RoverJames wrote: »

I can completely see how someone can just accept that the bookies are aiming to take 10 to 15% from bets and punt away selectively ignoring the value concept and just back horses (10, 20 ish a year) that they think have an awfully decent chance of winning.

Value is immaterial to them, the result is all they care about.

I'm different, I don't really care about the result, all I'm bothered about is laying at a price I consider to be poor value for the backer. If I do that 200/300 times a month I'm happy, I know 1 in 3 / 1 in 4 of them will win.

If I was only laying 20 horses a year the result would be far more important than the value side of things.

Top notch angle once again James, and portrayed with much more astuteness than I could have done so myself.

Glad to see that others can understand my view on this topic to an extent and not be affected by the red mist bandwagon that others succumbed to.

I must reiterate however that I have never said that I think the price of an animal or value is totally irrelevant as many have suggested.

Huntley wrote: »

I would take account of the price of an animal but to an extent I agree with you, it is too easy to get far too entangled with the value argument.

I have a different opinion on the importance of 'value' than the majority here but those who have suggested that I am "fundamentally incorrect" is a little ironic, one cannot be incorrect about a subjective topic.

The idea that a 'good' or 'bad' punter is solely based on their adherence to theories and statistics is far too literal. I would have great difficulty in accepting that a punter who can make money from the game could be classified as a 'bad' punter.

Colonel Sanders

RoverJames wrote: »

Again I'll repeat, for someone selectively backing 15 horse per annum, value isn't key, the result is. They just accept the bookie have taken their cut (many would expect all of their selections to be priced consistently "unfairly") and the punter ploughs on. They presume the price is no better or no worse than most of the other days prices and they feel they have an edge due to their knowledge or interpretation of the race and the other horses in it.

I'll repeat, someone merely concentrating on backing winners yet not getting value is more likely to result in losing money. If you are taking bad value bets but you make money over a finite time span this is a statistical anomaly or down to "chance" or variance

I've stated repeatedly that a "good" punter can end up losing money but a "bad" punter can end up winning money in the short term. Spin the coin once, the 4/6 punter might win and indeed outperform the 11/10 punter but is a very "bad" punter

A person might win the lotto twice over their lifetime but still be a poor gambler, in fact merely doing the lotto (as a gambling medium) is an indication of this

A roulette punter who can show a +ive RoI over his life is still a "bad" punter despite expecting to have a near 50% strike rate. It's pure chance (s)he managed to win via the medium of roulette

RoverJames wrote: »

It's very easy to speel on about value, in punting circles it's the done thing. It's not so easy to show positive ROI over a period, a few on here have done so, most haven't.

I'd like to know where I have said otherwise. In any case a +ive RoI is not alone an indication that the punter is "shrewd". The more the number of bets a +ive RoI is based on implies the higher the probability that the successful RoI is bad on "shrewdness" and not "chance" (this is the crux of the law of large numbers)

RoverJames wrote: »

All this talk of value and favourable odds is merely bluster unless you can demonstrate you have the capability to find value and favourable odds. It's like people talking about curling free kicks into the corner around the wall.

I have never stated I, or anyone else, are all singing all dancing greatest punter. What I have stated is that value is the fundamental principle of gambling.

RoverJames wrote: »

There are imo as few people backing selections that are value consistently as there are people with +ive ROIs

I don't actually understand what you're saying here. To show a +ive RoI over a long series of bets you must punt at value prices so I'd expect

"people backing selections that are value consistently" to be approximately equal to "people with +ive ROIs"

RoverJames wrote: »

Can you not concede there are other options, such as ...

3. The punters small number of bets (15) over the period showed they might well actually know more about racing than the majority of punters due to a high percentage of them winning.

The punter is either a "good" punter or a "bad" punter in the terms I've defined (regardless of strike rate). This is a fact

We cannot ascertain this credibly with a small sample size. Your 3 is actually a mixture of my 1 & 2. Plenty of people can pick the most likely winner but if they bet without the odds being in their favour in the long run they will lose

RoverJames wrote: »

I'm becoming increasingly bewildered at your reluctance to admit certain posters might be better at picking winners than the rest of us.

I have never said this may not be the case. what I have said that people picking winners with no regard for price are "bad" punters and any wins are statistical anomalies

I don't care if punter X can pick more winners than punter y, if punter X consistently places "bad" bets and punter y consistently picks "good" bets then punter y is "better", irrespective of strike rates

What I have said is that value is the key not picking winners

RoverJames wrote: »

You also haven't offered an approximation of what a large number would be, I asked would you consider 15 bets a year over 50 years a large number, the answer to that is Yes or No.

I have replied to this twice, and the answer is not yes or no. It's like asking will I win the lotto this year? The answer is not yes or no. The best you do is formulate the problem and approximate a probability the person will win or lose. Let me refresh your memory

The law of large numbers does not answer this question, indeed it was never intended to do so.

If by "large number" you mean "guaranteed to win" the number is infinity. The best you can do is say I'm happy to accept a probability of p% that I'll win money and try derive how many bets you need to place

In the example in the OP if you have 750 bets you win with probability 89.9%

If you only punt 4/6 shots that should be 1/2 then if you place 750 bets the probability you win is 99.9943%

Are these probabilities acceptable to your risk tolerance? If so then 750 is "large" if not then 750 is not

You seem to have read "law of large numbers" and focused on the 2 words "large numbers". The law of large numbers will not and was never intended to tell you what a large number is (nor have I claimed it would). Its application to gambling is:

1. over a long period of time your actual strike rate will get closer and closer to your expected strike rate
2. over a long period of time your actual RoI will get closer and closer to your EV. If you are a -EV punter the more bets you place the less likely it is you finish ahead due to variance.

In a limited set of circumstances you can derive the probability p% of winning given n bets but in most cases you would have to make a lot of assumptions and end up with a very approximate answer

To recap what I stated in my OP:

1. you cannot win money in the long run if you do not consistently obtain odds that under estimate your selections chances
2. A "good" punter can lose in the short term but the more "good" bets (s)he places the less likely it is they will show a -ive RoI
3. Conversely a "bad" punter can win in the short term but the more "bad" bets (s)he places the more likely it is they will show a -ive RoI

and this is why value is the key not merely picking the most likely winner irrespective of price

Colonel Sanders

Huntley wrote: »

I would have great difficulty in accepting that a punter who can make money from the game could be classified as a 'bad' punter.

What about a person who plays the same numbers each week on the lotto and manages to win the jackpot. This person has shown a profit playing lotto, I still think they're a "bad" punter

Ditto a lad walks into a casino and blindly bets on red on a roulette wheel. He walks out showing a profit, I still think that person is a "bad" punter

Colonel Sanders

Huntley wrote: »

I must reiterate however that I have never said that I think the price of an animal or value is totally irrelevant as many have suggested.

I have a different opinion on the importance of 'value' than the majority here but those who have suggested that I am "fundamentally incorrect" is a little ironic, one cannot be incorrect about a subjective topic.

You said the following:

Value has no importance if you are consistently picking losers

If you are consistently getting value the more "good" bets you place the greater the probability you end up with a +ive RoI

However, picking winning selections is the most fundamental aspect of successful gambling, not value as you seem to think.

This invalidates the law of large numbers

There is no mathematical fact which states that value is the most fundamental aspect to successful gambling.

The law of large numbers does assuming you want to win in the long run

This "mathematical fact" is a myth that piss poor gamblers like yourself cling onto in the hope of actually making some money some day. You can look at your statistical books and timeform all you want but on all known evidence here you have little to show for it apart from some mediocre attempt to look shrewd, when you are anything but.

This is pure toss

Do you understand the idiocy in stating that value is the most important aspect of successful gambling?

The only value that counts is a winning selection

so the 4/6 punter in the coin toss example got value and is a shrewd punter

Where is the mathematical fact that states value is the most fundamental aspect of gambling?

The law of large numbers

he only marker is the selection winning

Selecting horses that can win is far more important.

I determine value by the selection that is most likely to win.

What value is there in a losing selection? In my opinion, there is none.

You can just as easily lose in the long term by selecting overpriced horses.

All of the above is incorrect, while I do agree that one's perception of value is subjective

Colonel Sanders

No need to refresh my memory, I did not ask did the law of large numbers answer the question, also by large number I did not mean guaranteed to win, I actually don't know where you pulled that query from..........

Colonel Sanders wrote: »

The law of large numbers does not answer this question, indeed it was never intended to do so.

If by "large number" you mean "guaranteed to win" the number is infinity. The best you can do is say I'm happy to accept a probability of p% that I'll win money and try derive how many bets you need to place

I shall repeat myself and again request an answer, as I have mentioned I believe in value etc and understand completely how over a large number of selections it is important, here comes the question now CS.............

Is 15 bets per annum over 50 years a large number in your view ?



I firmly believe the law of large numbers can be applied to 15 roulette spins, 15 rolls of a die etc etc etc due to the repetitive nature of the event and the lack of external variance effecting the event and result.

I also firmly believe that a selective punter can identify a small number of betting opportunities over the course of a year where he feels a selection has a huge chance of winning, again this selective punter would be buying into the idea that the bookies are taking 10 to 15% out of the market and that some of his 15 selections may be bang on pricewise, some might be undepriced and some may be overpriced, he cares not one jot as he lets the bookies do their job and he does his

I've highlighted the question in bold for you, I'll also make the following points as you seem a tad confused as to what I was asking....

- By "large number" I do not mean "guaranteed to win", I mean is the integer in question considered a large number in your view, the size of the number is independent of the results.
- I am not asking does the law of large numbers answer the question, I am asking do you think 750 bets (on horseracing, UK and Irish), over 50 years is a large number.
- the answer is either YES or NO or I am not answering

Colonel Sanders

RoverJames wrote: »

Is 15 bets per annum over 50 years a large number in your view ?

I am assuming by "large" you mean it's large enough such that the probability a good punters returns a +ive RoI is quite high

If this is the correct interpretation then in my view 750 bets is "large". So "Yes"

RoverJames wrote: »

the answer is either YES or NO or I am not answering

FWIW it is not a simple Yes or No question, answering "Yes" or "No" is subjective, similar to me asking will the lotto be won next Saturday

Interesting, cheers for the answer

Colonel Sanders

RoverJames wrote: »

Interesting, cheers for the answer

Apologises for the pedantry

I think it's a symptom of doing a pure maths degree

Colonel Sanders

Colonel Sanders wrote: »

................

I think it's a symptom of doing a pure maths degree

No bother at all, you might be right :pac:

The old one about the mathematician maintaining a battery is never fully charges whereas the scientist considers the battery fully charged after the charging curve levels out (sort of like radioactive decay and the 5 half lifes carry on)


I've mentioned already anyway but I'd consider a lifetime punting total of 750 bets over 50 years as very small due to there being 10,000 + ish races in the UK and Ireland per annum.

I reckon if a chap is that selective that he's operating in a somewhat stratospherical level and at a level of activity that wouldn't lend itself to an outsider making any sense of his methods.

Wouldn't be cut out for it myself tbh

Although I haven't had a decent bet on a horse since the Christmas period and don't expect to have any soon either.

dave3004

I read (I think it was RJ) who said value is immaterial to bookies and the result is all that matters. Just to let you know. Thats wrong.

We are as value based as the punter is. Maybe more so.

The only way to beat the bookies is to pick the value in their markets. The only way you can pick value if is there is some to pick.

Quick Example for ya.

If Joe Bloggs has 5 bets, all 10/1 pops. He puts 1k on all of them on the nose. They all win at 12/1. Hes taken 50k off the bookies.

Joan Briggs has 5 bets all 10/1 pops. She puts 1k on all them. They all lose at an SP of 4/1.

Joan Briggs account will be closed down before Joe Bloggs. Its not the result that matters to bookies, its value & shrewd punters.

If the punters business looks shrewd we will exclude them. But we still allow people who are beating us to keep betting cause we know we will get them in the long run.

Dont know if its been mentioned before but bookies have price marks too which help us remove value from markets. So lets say Joe Bloggs is a shrewdie. We only let him on to win 1k in any race. I go up early market on a race in Newbury and he backs a 20/1 pop. Once I see that bet in my screen (it will pop up coz hes shrewd). I will turn off the price a few rolls. Into 16s or so. Now anyone else who fancies it will be getting paid 16 to the euro and not 20. So although Joe Bloggs will beat the bookie consistently, he saves thousands for us in every race and his business is worthwhile.

Excellent thread guys.

RoverJames

RoverJames wrote: »

I'm a firm believer in value and tissue prices etc, I aim to lay about 3000 to 4000 short priced horses in 2013 and make a decent wad of cash in doing so. We all know if you lay 3000 to 4000 horses on the exchanges willy nillly you'll lose your arse.

I can completely see how someone can just accept that the bookies are aiming to take 10 to 15% from bets and punt away selectively ignoring the value concept and just back horses (10, 20 ish a year) that they think have an awfully decent chance of winning.

Value is immaterial to them, the result is all they care about.

I'm different, I don't really care about the result, all I'm bothered about is laying at a price I consider to be poor value for the backer. If I do that 200/300 times a month I'm happy, I know 1 in 3 / 1 in 4 of them will win.

If I was only laying 20 horses a year the result would be far more important than the value side of things.

dave3004 wrote: »

I read (I think it was RJ) who said value is immaterial to bookies and the result is all that matters. Just to let you know. Thats wrong..............

You misread my speel

The "them" I was on about was the backer, not the bookie

I crudely reference the 10 to 15% as the bookies cut, I know it's often more and often less but in guesstimate land 10 to 15% does me fine

As a layer, as I mentioned above the result isn't a concern to me, only the price I lay at.

I'm going to stick my neck out a bit, I've about €1500 in my betfair account, built up from €200 last August, by laying and laying alone I'm hoping to get that to €10,000 + by end of 2013, mainly at the weekends. Value is all I worry about in that sphere. I have been posting my results after every 600 lays in the gambling forum and will continue to do so.

Slattsy

RoverJames wrote: »

I'm going to stick my neck out a bit, I've about €1500 in my betfair account, built up from €200 last August, by laying and laying alone

Impressive.

I've never laid a horse in my life (i've laid a few hounds mind you ) but its something i've often thought about but never get around to.
I may make a concerted effort to lay the odd one soon though.
I'll try find that log of yours and look for some advice and how you pick them.
Good luck getting to 10k.