Measuring / Predicting Success

keith16

Having a read of the 6 Nations review thread got me thinking on the how subjective Rugby can be sometimes. Obviously, from watching the tournament Mike Brown was one of the standout players. But I would like to know, statistically, how much better than everyone else he was...and why?

At a team level, what are good indicators of potential success? Now, I as a rugby fan obviously know that things like tackle completion are good indicators, but there are probably a lot more I am not aware of.

Ball won in opponents 22 is always one I see, but I'm not sure what that means in reality.

Full disclosure....I am attempting to do a statistical analysis (notice the link in my sig re data) and would like to get some fans insights on the type of data to collect and the types of questions to ask of the data.


Thanks!

chris85

keith16 wrote: »

Having a read of the 6 Nations review thread got me thinking on the how subjective Rugby can be sometimes. Obviously, from watching the tournament Mike Brown was one of the standout players. But I would like to know, statistically, how much better than everyone else he was...and why?

At a team level, what are good indicators of potential success? Now, I as a rugby fan obviously know that things like tackle completion are good indicators, but there are probably a lot more I am not aware of.

Ball won in opponents 22 is always one I see, but I'm not sure what that means in reality.

Full disclosure....I am attempting to do a statistical analysis (notice the link in my sig re data) and would like to get some fans insights on the type of data to collect and the types of questions to ask of the data.


Thanks!

The RBS 6 nations site itself has a bunch of stats available for each game. There are heaps of metrics looked at and some are more important than others depending on the position being looked at.

razorblunt

Give him plus points for the "save" on BOD's kick through and minus points for the sly jab on Huget. Please and thanks!

GiftofGab

Scrum.com is the best site for stats. I not a huge fan of stats to be honest. They can be sometimes be misleading. They dont measure alot of important issues in a game of rugby.

If I was going by stats the indicators of success I would at would first of all break down each players positions and role individually then see how well they performed their role. For example, how many scrums a tighthead lost / won. The scrum and lifting in a lineout is a tightheads job - anything else is a bonus. Also how many lineouts a 2nd row won / lost along with their tackle count. Back row - how many turnovers he won along etc.

But nothing beats the naked eye watching a match. You would have a good idea who had the best impact on a match simply by looking at it without relying on stats.

CatFromHue

I don't think you can statistically show Brown is better than Sexton or Parisse.

Stats only show you one aspect and you have to use them to compliment your opinion.

Ireland have only won 5 out of 8 games under Schmidt which is only 63%. That doesn't take into account who we played or how the games actually went. You could argue that the way we played against NZ and Eng would actually be a plus for Schmidt and not a negative.

pickarooney

Demented Mole has some scarily in-depth statistics which he uses to evaluate the contributions of forwards.

keith16

razorblunt wrote: »

Give him plus points for the "save" on BOD's kick through and minus points for the sly jab on Huget. Please and thanks!

pfft....other way around tbh :pac:

GiftofGab wrote: »

If I was going by stats the indicators of success I would at would first of all break down each players positions and role individually then see how well they performed their role.

But nothing beats the naked eye watching a match. You would have a good idea who had the best impact on a match simply by looking at it without relying on stats.

That's interesting....but aren't there a lot of ways you could evaluate players performance based on core skills...carries, yards made, men beaten, handling errors, tackles made etc? I think there are definitely attributes you can assign to players and then perhaps add a weighting based on your position (higher weightings for defenders beaten for wingers for instance).

I just think that reputation and naked eye can sometimes be misleading, everyone is biased after all.

pickarooney wrote: »

Demented Mole has some scarily in-depth statistics which he uses to evaluate the contributions of forwards.

Who now?

who_me

(Kind of stating the obvious...) there is no such thing as "better". The metrics you choose to collate and the weighting you give to each will determine the "best" player as much as their performance itself.

Most would agree BOD was one of our finest ever players, but could a prop ever be our best ever player? Does the fact that he won't make many line-breaks, or score many tries out wide, or kick any goals mean he can't be viewed as a great?

It's like trying to scientifically prove ice cream is better than chocolate. I guess the imprecise, unscientific, arbitrary-ish metric we tend to use when judging players is: "Is he/she doing something positive that other players (on his team or the opposition) rarely/never do?". Linebreaks / exceptional kicks / turnovers etc. I think it's more difficult for someone like a prop to stand out in most statistical analyses, as his chief contributions (scrum, lineout, maul) are all collective achievements, not individual.

durkadurka

Here's the demented mole with an extraordinary statistical analysis of the lions tests:


http://dementedmole.com/2013/10/30/ruck-marks/#more-3681

CatFromHue

Second Captains did a segment on Stats in sports couple of weeks ago that was interesting enough.

http://www.irishtimes.com/sport/second-captains

It's the "Second Captains 20/03 - Moyes hope, Schmidt stats, rise of..." one.

It starts after 4 or 5 mins

Prawo_Jazdy

I've often wondering aswell how susceptible people are in taking in the commentaters emphasis on certain players in making their judgements. I'd reckon if you done a test on two groups of people, one watching say rte and one watching bbc for a match , there's a good chance who people say stood out might reflect who the commentaters thought did.

Not that they'd differ completely, but I often feel certain commentaters go absolutely mad when certain players just touch the ball, while overlooking others that have a great game .