Matter of Stats

View Original

2016 Round 24 (Finals Wk 1) : No Real Surprises

Since the turn of the century, upsets have been relatively rare in Finals, so with the four home teams all mild to firm favourites this week, it's as well that the Head-to-Head Fund is constrained to betting on home teams alone.

As it turns out, that Fund liked the prices that were on offer for both the Cats and the Crows when the markets first opened, which is why Investors find themselves with the standard-sized bets on those two games.

The Line Fund also saw a little value in the same two games and, accordingly, made a couple of smallish wagers of its own, also on the Cats (-6.5) and Crows (-30.5).

Also, in the overs/unders markets, the 164.5 total on offer in the Eagles v Dogs game, and the 177.5 total on offer in the Crows v Roos game were both assessed as providing sufficient edge to induce the Overs/Unders Fund to take the plunge in both cases.

In total, the six wagers spanning three of the weekend's four contests represent 3.6% of the original Overall Portfolio and have a maximum upside of 2.5%. The detailed risks and rewards from the head-to-head and line bets are summarised in the ready reckoner below.

TIPS AND PREDICTIONS

Surprisingly, the only dissension amongst the Head-to-Head Tipsters has come this week, not in the game with the smallest line market handicap (the Cats v Hawks game), but instead in the Swans v Giants game, which has a handicap about 2.5 times as large.

In that game, it's C_Marg and ENS_Linear_MoSS that have opted for the upset Giants victory. Interestingly, MoSSBODS_Marg itself has tipped a small Swans win in this same game, cannily mirroring BKB's tips and therefore preserving its one tip overall season lead.

(By the way, this season I've carried HSH and CTL into the Finals. Their heuristics, I think, flow fairly naturally into this part of the season, HSH selecting whichever is the designated home team and CTL whichever of the teams in a contest finished higher on the ladder at the end of the home and away season.)

It's the Swans v Giants game that has also created the most disagreement amongst the Margin and Head-to-Head Probability Predictors. 

Amongst the Margin Predictors, the average mean absolute deviation (MAD) has come in at just over 7 points per Predictor and the range of predicted margins spans 4 goals. For the Probability Predictors, the mean MAD is over 8% points per Predictor in this game, and the probability assessments span 21% points, from C_Marg's low of 47% to Bookie_LPSO's high of 68%.

MoSSBODS DETAILS AND MoSSBODS COMPARISON WITH THE TAB

Firstly, a look at MoSSBODS' component team ratings and venue effects reveals that all four of the (putative) home teams would have been favourites even without their home ground advantages.

In fact, in the Swans' case, the choice of venue is slightly to their detriment and roughly halves what would otherwise have been their predicted margin of victory according to MoSSBODS.

As is fitting for the generally lower-scoring nature of Finals, MoSSBODS has only the Crows topping the 100-point mark this week, and has three of the four contests generating fewer than 175 points.

Comparing MoSSBODS' opinions with those of the TAB Bookmaker shows broad consensus, the Bookmaker also predicting only the Crows to score more than 100 points and that three of the four games will produce fewer than 175 points in total.

Still, in the Eagles v Dogs game his team and total projections are such that MoSSBODS feels an unders wager is justified, and in the Crows v Roos game those projections are such that MoSSBODS has signed on for an overs wager.

In the final round of the home and away season the TAB Bookmaker outforecast MoSSBODS in terms of home team scores, total scores and game margins, leaving MoSSBODS to prevail - as it has for much of the season - only in terms of away team scores.

That leaves the overall performance statistics for the season to-date as follows:

Home Team Scores

  • MoSSBODS: 19.7 MAE; +5 Average Error (Actual - Predicted)
  • TAB: 17.7 MAE; +2 Average Error

Away Team Scores

  • MoSSBODS: 20.3 MAE; 0 Average Error
  • TAB: 20.5 MAE; -5 Average Error

Aggregate Scores

  • MoSSBODS: 26.9 MAE; +6 Average Error
  • TAB: 25.5 MAE; -3 Average Error

Game Margins

  • MoSSBODS: 29.9 MAE; +5 Average Error
  • TAB: 29.1 MAE; +6 Average Error

FINALS PERFORMANCES BY LADDER-FINISH 2000 TO 2015

Though they represent only a small sample, the combined results from the last 16 Finals series still tell an interesting story, especially in relation to Week 1 of the Finals.

Across those 16 seasons, the Minor Premiers and Runners Up have both won 12 of their 16 Qualifying Finals, 6th-place has defeated 7th-place in 11 of their Elimination Finals, but the 5th v 8th matchup has been split 8 all.

Combining the results from all weeks of the Finals, the Minor Premiers and Runners Up have by far the highest winning rates, but it's the teams in 3rd and 6th that have the next-highest rates of around 50%. Thereafter, the teams in 4th, 5th and 8th all have roughly the same rates (36%), albeit that the teams in 4th have progressed to the Preliminary Final or beyond far more often.

Teams finishing 7th have the worst collective record, winning less than one-quarter of their Finals and failing to progress past the Semi-Finals even once.