TEN YEAR TRENDS

tomaussie

Hi guys.

I'd like to get your opinions on ten year trends.

I posted a few times recently on the issue and I'm wondering what other people think.

I suppose I believe that History does usually tend to repeat itself.

So based on that.... what races do you think trends are more relevant for ?

I'd imagine that large handicaps are races whereby trends would be relevant ?

I recently dissected two races based on trends (posted by your good selves upon my request) and the winners made the final 3 (by final 3 i mean that it was one of 3 horses to make the cut that ticked all the trends) in one race and the final 6 (from 28 runners) in another.

So there seems to be some validity to use trends based on my very limited experience.

Any thoughts are welcome. I may be missing out on specific questions regarding ten year trends but I'm sure you guys will steer this thread in the right direction.

<snip>

Looking forward to your comments guys.

abarkie

Hi

About 5 or 6 years ago the Grand National was a "trends" race

Handicap mark between 147 and 157 (from memory)

Won over 3m+

Nothing more than 11 stone

Then they amended the fences and the handicapper compressed the weights and the rest is history

Anyway, the Irish National is as below from 2010:

KEY Irish National Trends (This left shortlist of 4 last season inc. Niche Market)

10/10 - No More Than 8 Runs in Handicap Chases
10/10 - No more than 13 Chases
10/10 - No more than 3 runs that calendar year (8/8 - No more than 2 runs)
9/10 - OR between 125 + 136
9/10 - 10-12 or less (9/9) (28/30 placed horses 11-0 or less)
9/10 - 7-9yo's
9/10 - Won over @ least 3m

*Course form a positive for Irish-trained winners

Haven't looked at it much since.

Hope this helps

Barbaric Gentleman

Almost all trends are worthless. Talk to anyone with a basic knowledge of statistical analysis and they will tell you the same. It's a lazy approach.

j@utis

I don't like the trends based on winners only because the data pool is too small. Even over 10 years you only have 10 data points to analyse - sorry but you can't make any meaningful statistics out of that. e.g. abarkie's mentioned GN winner should be 7-9yo, but what if all runners in the race fall into this category? in this case, of course, the winner will come from this age group - so age is factor is out. the same can be said about weight carried etc.
what I'd like to see is comparison of the winners to the rest of the fields. it's a lot of work to take all imaginable characteristics of the all runners, group them, sum them or do some other magic with the data and see which of them could qualify as a decisive factor.
I tried something like this couple of years ago for one of the big races in Galway and I had 3 of the first 5 home, including the winner.

NewApproach

Ten year trends are irrelevant. Totally insignificant sample size.

Clonsuttonlad

I think trends are useful as it helps you form a profile for the type of horse that has won the race.

I find trends are particularly useful for flat races. Draw trends are vital.

tomaussie

Thanks so far guys. Looks like I might be barking up the wrong tree.

When I analysed those two races previously using the 10 year trends I felt there may have been something in it as the winner of the Melbourne cup was a stand out and the other was johns spirit who made the final 6 of the cut.

I thought there was something to it.

I'll do it again for 2-3 more races (suggestions as to what race or types of race are welcome) and see how I get on.

I agree especially with the lazy comment above but only because that describes me pretty well so you may be right.

abarkie

I think it might help you to narrow a field down

Years ago I remember reading a Timeform article stating that something 80% of all races are won by the top 2 in the ratings

To my mind, its a combination of things that should help you form your opinion and 1 of them is a race trend - if indeed there is one

NewApproach

NewApproach wrote: »

Ten year trends are irrelevant. Totally insignificant sample size.

On a purely mathematical basis you are right - nobody should back on trends from 10 renewals of a race.

On the other hand, trends are a symptom and not a cause. They are usually just a short hand for a process of elimination a reasonably intelligent punter might get around to anyway. If you can see the 'reason' why a trend exists, by all means follow it.

tomaussie

So is there a consensus as to what types of races where ten year trends may be more relevant ?

For no good reason I'm thinking big weekend handicaps ?

Clonsuttonlad

Big field races on the flat, particularly sprint handicaps where sometimes you can rule out a large number of the field based on draw number/weight/age.

However trends are always useful as a guide in my opinion. They give you a feel for the race and the type of horse that usually wins. They should not be applied absolutely though. Whenever trends are available it's worthwhile having a look at them. I think people try to make racing an exact science, which it's not. A perfect statistical sample over 100 or more years would be less useful than studying ten year trends.

ste2010

tomaussie wrote: »

So is there a consensus as to what types of races where ten year trends may be more relevant ?

For no good reason I'm thinking big weekend handicaps ?

I could think one reason not. At least half of the big handicaps won on Saturdays are won by plot horses. It would be the non handicaps that I'd use trends for more so. To be honest I've posted something similar about obtaining information to do deep analysis before. Where you measure everything from age progression, ground, hurdle/fence progression by age, by race count, breeding, speed. I honestly believe its possible to become more scientific on the information available that's currently being done by software etc. the big question is would it be worth the analysis, would it give you the edge and for the example I mentioned above re plotting. In handicaps it wouldn't because there's a ghost in the machine, a rat in the kitchen, a worm in the apple or an Jonjo at a racecourse :-)

ste2010

ste2010 wrote: »

I could think one reason not. At least half of the big handicaps won on Saturdays are won by plot horses. It would be the non handicaps that I'd use trends for more so. To be honest I've posted something similar about obtaining information to do deep analysis before. Where you measure everything from age progression, ground, hurdle/fence progression by age, by race count, breeding, speed. I honestly believe its possible to become more scientific on the information available that's currently being done by software etc. the big question is would it be worth the analysis, would it give you the edge and for the example I mentioned above re plotting. In handicaps it wouldn't because there's a ghost in the machine, a rat in the kitchen, a worm in the apple or an Jonjo at a racecourse :-)

But some of the trends are there to help you spot the plot.

Off the top of my head, there's a strong trend for the Fred Winter - exactly three hurdle starts, no winner first or second time out - that helps narrow it down to the horses that have been campaigned with the race in mind and to do the bare minimum to get a (not particularly troublesome) handicap mark.

yaboya1

Deleted User wrote: »

But some of the trends are there to help you spot the plot.

Off the top of my head, there's a strong trend for the Fred Winter - exactly three hurdle starts, no winner first or second time out - that helps narrow it down to the horses that have been campaigned with the race in mind and to do the bare minimum to get a (not particularly troublesome) handicap mark.

Although that has been a pretty reliable trend for the Fred Winter, you still get horses that don't fit the mould but still win. Four of the last five don;t match that stat if I remember correctly (and the race is only in existence since 2005!). I think you should use trends as a rough guide, but try and find the reasons why such trends exist rather than assuming they will automatically repeat themselves.

yaboya1

yaboya1 wrote: »

Although that has been a pretty reliable trend for the Fred Winter, you still get horses that don't fit the mould but still win. Four of the last five don;t match that stat if I remember correctly (and the race is only in existence since 2005!). I think you should use trends as a rough guide, but try and find the reasons why such trends exist rather than assuming they will automatically repeat themselves.

agree 100%. Didn't know that about 4 of the last 5 fred winters... so much for the top of my head...

yaboya1

Only reason I know is because I was on most of the horses that fit the stat and the winners didn't! :eek:
Nice touch on Crack Away Jack at 14/1 the year he beat Ashkazar though

ste2010

[Deleted User] wrote: »

But some of the trends are there to help you spot the plot.

Off the top of my head, there's a strong trend for the Fred Winter - exactly three hurdle starts, no winner first or second time out - that helps narrow it down to the horses that have been campaigned with the race in mind and to do the bare minimum to get a (not particularly troublesome) handicap mark.

That's rare. What Im saying is, for big Saturday or most handicaps that trends mean nothing if there's a Jonjo horse which has been purposely running flat

ste2010

ste2010 wrote: »

That's rare. What Im saying is, for big Saturday or most handicaps that trends mean nothing if there's a Jonjo horse which has been purposely running flat

But that's a trend! It's such a well understood trend that Jonjo / JP horses go off favourite for just about every grade 3 handicap out there.

In that case it would be "Jonjo/JP horses have won 3 times and been placed twice from 9 total entries in the past 10 years, respect any entry with these connections and be prepared to look beyond recent form"

ste2010

That's wrong...Jonjo horses don't always off favourites in G3..that's light analysis.
I'm assuming the OP is looking for something that stands up in terms of gathering data. When you do a detailed level of analysis you want to have as much of it automated as possible. Then Apply your manual angle such as this one or when money comes. The possibilities are endless. My point is unless you factor everything in for handicaps then trend analysis is worthless

ste2010

ste2010 wrote: »

That's wrong...Jonjo horses don't always off favourites in G3..that's light analysis.

It's not 'analysis', it was an exaggeration for effect (although I didn't say 'always'). The point is that Jonjo's record with JPs horses is well known and punters latch onto anything entered, and why not. My other point is that this is what is known as a TREND. They outperform against the field in these races and thus a trends punter will look closely at their merits.

ste2010 wrote: »

I'm assuming the OP is looking for something that stands up in terms of gathering data. When you do a detailed level of analysis you want to have as much of it automated as possible. Then Apply your manual angle such as this one or when money comes. The possibilities are endless. My point is unless you factor everything in for handicaps then trend analysis is worthless

Trend analysis, if done right, can usually cut down the field considerably for minimal effort. It doesn't find you a winner, I don't think anyone would ever claim it did, but it is not 'worthless'.

Most punters use trends anyway, even if they think they don't.

Harry Angstrom

There was a time, up until a few years ago, when if you wanted to pick the Grand National winner, you backed a horse carrying no more than 10-13. From 1989 to 2008, only one winner had carried 11 stone or more to victory. Winners are coming from all over the shop since then though, with the bias now strongly turning towards horses carrying 11 stone and above. Looks like that earlier stat bubble has burst.

tomaussie

Anybody got a link to the trends for the LADBROKE HURDLE ?

I'm going to try again.