NRL Positions of Power

TipBetPro is a player based statistical model used to predict the outcome of NRL, AFL and shortly cricket matches (first match is England v Sri Lanka next week, and look out for @TBP_Cricket for views and news).

Can we improve?

So over the season to date we have been asked a number of questions and been provided with some great insights into how we can improve on our predictions.

For those uninitiated with the TBP method, we basically calculate the impact every player has on a game (using analysis of historical statistics), add them up for each team, and take the difference. We then factor in team based factors, such as home ground advantage, ladder position and form. The combined difference is our predicted margin for the week….and viola there we have it.

Does position matter?

Now in terms of the player ratings we simply add up all 17 players scores and do not directly adjust for any positions or whether on the bench or not. We have our reasons!

But do certain positions influence a game more than others? Well you’d think so, wouldn’t you?

So we recently held a survey on twitter (@tipbetpronrl) which asked what you thought on positions and their impact on the results:

So, the public has voted and it’s the Spine (1, 6, 7 and 9) which has the biggest impact on the result. This could be interpreted as saying the team with the better Spine will win more games, although you obviously can’t ignore the other 13 players!

Does a straight back wins you games?

Interestingly, based on TBP rankings the best 17 named on Tuesday this week (Round 10), was North Queensland Cowboys, followed by the Sharks, and then the Broncos. But based on the Spine alone we see a different (ish) ranking:

Melbourne jump up to top spot! And you can see our favourites, the Dragons are punching above their weight in 5th spot (for the Eels, we know Peats has left, but he hadn’t on Tuesday)!

But remember we don’t distinguish between the Spine and the other 13 players. Should we?

A bit of grouping

Well to try and apply some rigor this we have gone back to every game in rounds 1-9 and calculated the following:

  1. The TBP rankings for the backs (1-8)

  2. The TBP rankings for the forwards (9-13)

  3. The TBP rankings for the spine (1,6,7 and 9)

  4. The TBP rankings for the bench (14-17)

  5. The TBP rankings for the team (1-17)

In the first instance, we have then simply calculated the difference for each game and for each of the five groups and then seen if the team with the higher rankings won.

Now before we show you the results, there have been 72 games in the first nine rounds, of which we have tipped 46 (64%) which compares to 47 favourites (65%), so we are one behind the favourites (although not after the Dragons won Thursday night!)

So without further ado:

Well this tells us three things:

  1. The bench is important – very

  2. The Spine is more important than forwards and backs - just

  3. If we just used player rankings we’d have 51 tips right, and NOT 46! Maybe we should ignore ladder position, form and where the game is being played!

The above analysis is a bit misleading as you can’t just isolate one part of the team and ignore the other parts, so of course the “team” factor will always be better. But if we adjusted or weighted some of the groups could we improve on 51 correct tips?

Well yes we can – in fact if we weighted some of the positions we could get to 55 correct tips

Now off course this is all well and good when we know the results and is clearly not statistically valid, but it is something we will need to test further to refine our approach.

In conclusion, we are happy to continue rating all 17 players and we believe it's further evidence is a team game, not an individual one!

Keen to hear your thoughts.

GUEST WRITERS WANTED – Do you have a view on anything NRL related? Maybe a response to this article or something else. We are happy to post here* and make you famous (well sort of).

To see our betting recommendations, including our Bet of The Week, and to read more about our approach, and to compare our performance with your own, please visit www.tipbetpro.com or follow us (abuse us) on Twitter @tipbetpronrl.

*Subject to editors discretion


Featured Posts
Recent Posts
Archive
Search By Tags