Wallabies 2012

[boyo]

Well done on the draw with the NZRT last night. The ball skills need a lot of work.

Although they won the series with Wales 3-0 I think that the winning margins should have larger.

Now for other opinions.

 

[boyo]

Oh well, it appears that few people read this. The success or otherwise of "Suggested threads", or my thread topic, remains to be seen.

 

[daz]

Don't take it to heart, boyo.

The topic idea is just too large for a single thread. There are many other Wallaby related threads where your idea could fit in so this thread is probably redundant.

Thanks, and don't stop!

 

[Schadenfreude]

I can't create a new thread - or suggest one - so I'll just tack this on here.

I watched Moneyball the other night, and it looks to me like the Wallabies coaching staff may have fallen in love with the concept a bit too much.

The idea when buying players is to buy points, and defence.

A player like McCabe becomes valuable, because he creates more points than he gives away.

A player like Cooper, in a world of increasingly impenetrable defences becomes less valuable.

Pocock creates penalties, and slows the attack, more valuable.

The more I think of it, I'm sure that's what they're doing.

Perhaps one day there would be a forum that would let me create a post about it.

 

[en_force_er]

I can't create a new thread - or suggest one - so I'll just tack this on here.

I watched Moneyball the other night, and it looks to me like the Wallabies coaching staff may have fallen in love with the concept a bit too much.

The idea when buying players is to buy points, and defence.

A player like McCabe becomes valuable, because he creates more points than he gives away.

A player like Cooper, in a world of increasingly impenetrable defences becomes less valuable.

Pocock creates penalties, and slows the attack, more valuable.

The more I think of it, I'm sure that's what they're doing.

Perhaps one day there would be a forum that would let me create a post about it.

I really doubt that. Rugby Union has far too many variables and I doubt any successful coach would use stats (and probably not the ones listed above) as anything more than a guide line.

On top of that, does RDeans seem particularly progressive in his thinking to you?

 

[Schadenfreude]

I really doubt that. Rugby Union has far too many variables and I doubt any successful coach would use stats (and probably not the ones listed above) as anything more than a guide line.

On top of that, does RDeans seem particularly progressive in his thinking to you?

Remember 10 man rugby? That was 100% based on stats.

This is the only theory that I've come up with that explains what we've seen from Deans.

The "it's too complicated" argument is the same one they trotted out in baseball.

I'm not saying I agree it's the right way to do things, but using this model would make sense of a lot of things we've seen.

 

[en_force_er]

Remember 10 man rugby? That was 100% based on stats.

This is the only theory that I've come up with that explains what we've seen from Deans.

The "it's too complicated" argument is the same one they trotted out in baseball.

I'm not saying I agree it's the right way to do things, but using this model would make sense of a lot of things we've seen.

Look, I know you think you're on to something but be logical. The idea Deans is basing what he's seen off a baseball system that even still isn't completely mainstream within basebase in pretty far fetched.

Baseball is a sport with a far smaller amount of variables to rugby. In fact, rugby probably has the most variables of any mainstream sport. Did Deans select McCabe because he valued gain line metres in a 12 more than flash? Yes, playing conservatively isn't a new theory.

Is this because of somebody is a lab coat? Are we to blame Italian Soccer's conservative style that's existed for probably 50 years on the quite recent Moneyball system?

 

[Schadenfreude]

It's not a baseball system. It's mathematics/statistics/economics.

 

[en_force_er]

It's not a baseball system. It's mathematics/statistics/economics.

It's called Sabermetrics (Society for American Baseball Research), the system was designed for the very statistical baseball.

Perhaps sports like cricket could use it a bit more, with rugby stats can only be an influencing factor because there's just too many. Pick key ones for sure, but if you want an algorithm to work out how good a player is you'd need a pretty crazy formula, from there you'd need to ensure that stats gathering is on par (which in rugby it often isn't).

 

[Schadenfreude]

There's no reason it can't be applied to Rugby.

Obviously the calculations change, and there are some other things to consider. But if you think statistical analysis isn't used you're out of your gourd.

So how far does the analysis get used? It should be for game planning. May be used for team selection. But is it used for contract negotiations?

 

[en_force_er]

There's no reason it can't be applied to Rugby.

Obviously the calculations change, and there are some other things to consider. But if you think statistical analysis isn't used you're out of your gourd.

So how far does the analysis get used? It should be for game planning. May be used for team selection. But is it used for contract negotiations?

I didn't say it wasn't used, I said there's no cover-all formula or even anything close like you've previously expressed.

There's a different between saying "this openside flanker is an asset to our team because he effects X amount of rucks, is a good carrier, and a good leader" and a maths formula.

One is a combination of statistics and logic, whereas the other would be the sabermetric style of think you've spoken of.

Some sports (read: most of the American ones) are incredibly easy to quantify in statistics. Australian sports, especially rugby, are not.

 

[Schadenfreude]

You might be more wrong than you think.

I'm not convinced it would work, I am convinced it would be worth examining closer.

 

[TOCC]

Most Super Rugby clubs have a small team of analysts crawling over a game replay in the hours following to piece together a post match report.

I imagine the likes of the RFU invests far more heavily into this more then other nations or teams would, but i have also heard Link mention the performance of a player using statistics as the unit of measurement so i'm not entirely sure its reserved for just the test level..

Yes rugby union would be significantly more difficult then the likes of Baseball but it is definitely still applicable, you only need to look at the Foxsports Fantasy sports to see how this could be measured.

 

[Richo]

The difference is that baseball lends itself to modelling. First, the game itself is very simple in structure with discrete elements that can be analysed independently. Second, the baseball season is 162 games long and each player appears multiple times at bat (or on the mound) in each game. That means sample sizes become large enough to be reliable very quickly.

Rugby is completely different. I'd think of it as a lot more like basketball, which is also difficult to model because so much changes with every 'event' on the court. Even American football is somewhat simpler, because of the discrete plays. Rugby is also affected by variables like pitch condition and referee style which further impact the reliability of sample sizes. RuckingGoodStats has written some good posts on this that I don't have time to dig up.

I have no doubt that teams are using a lot more statistics than we know about, but I highly doubt it approaches anything like the Moneyball model. More likely, teams track a lot of basic and advanced stats, some might even have advanced evaluation stats like baseball's VORP or basketball's PER, but this would be strongly balanced by subjective player-ratings, preferences for playing style, existing personnel (i.e. if you have Quade Cooper, you play an attacking, countering game), and so on. In other words, I would expect most teams emphasise traditional scouting and statistics, supplemented by statistical valuations as a way of reality-checking what they see, or to uncover hidden value. There are areas in which statistic would be extremely helpful, such as tracking success rates of kickers based on field positions. Captains probably learn those probabilities and use them to help make decisions about place-kicking.