2017 - Round 24 (Week 1 of the Finals) : A Bye Then The Tigers?

/

With the bookmaking markets posted, all of the MoS algorithms that use prices as inputs have what they need to do what they do, so, the bye week notwithstanding, there's no reason not to post the full set of forecasts today.

And in those forecasts, the only dissenting voices amongst the Head-to-Head Tipsters are the MoS twins who've both tipped the upset in the Cats v Tigers clash.

With the two RSMP Tipsters opting instead for the Cats in this game, we'll finish the round either with a three-way tie between those two and MoSSBODS_Marg at the top of the Leaderboard, or with a two tip lead (and only five games to play) for MoSSBODS_Marg.

Amongst the Margin Predictors, by far the most disparate set of forecasts have come in the Crows v Giants game where C_Marg and the MoS twins have the home team winning by about 5 goals while the rest of the field has the margin for the Minor Premiers at about half that.

In the other matches, we have C_Marg and MoSHBODS_Marg defining a three-goal range in the Cats v Tigers game, Bookie_9 and RSMP_Weighted defining a 10-point range in the Swans v Dons game, and C_Marg and RSMP_Simple defining a 2-goal range in the Power v Eagles clash.

MoSHBODS_Marg now trails RSMP_Simple by about 13 points, and will see that gap shrink or not depending on the outcomes of each game roughly as follows:

  • Reduce by 17 points if Adelaide win by 33 points or more
  • Reduce by 11 points if Richmond win by 4 points or more
  • Reduce by 7 points if Sydney win by 30 points or less
  • Reduce by 1 point if Port Adelaide win by 15 points or more

The first two games then are the crucial ones in the context of the Margin Predictor race for (desperately insignificant) glory.

The Head-to-Head Probability Predictors also have the most disparate probability estimates for the Adelaide v GWS clash, those estimates spanning a 21% point range from Bookie_OE's 63% to MoSHBODS_Prob's 84%.

C_Prob is at the high end of the probability estimates for the home teams in every other game.

MoSHBODS_Prob, the current leader, will suffer most if Adelaide lose.

WAGERS

MoSHBODS' confidence in Adelaide sees Investors with a very sizeable (okay, let's call it 'large') head-to-head wager on them, and MoSSBODS' similarly positive view sees them with line wager as well, albeit a less startlingly-sized one.

The only other wager in these markets is a small head-to-head bet on the Tigers.

In total, the three wagers represent 2.5% of the Overall Portfolio and, together with the sole over/under wager that we'll review in a minute, mean that every dollar in that Portfolio has now been put at risk more than twice during the course of the season.

Unsurprisingly, our risk this week is concentrated in the fate of the Crows, a win by them of more than 14 points capable of adding 1.5% to the Overall Portfolio, and a loss capable of subtracting 2.3%.

Richmond can do far less good and considerably less damage. The difference between a best-case and worst-case result for them will only move the Overall Portfolio by 0.4%.

I expected the head-to-head and line markets would go up sometime this week, but was surprised to see the unders/overs markets go up on Wednesday as well, but up go they did, the details of which are below.

Based on those markets and the opinions of the MoS twins, the Geelong v Richmond game is universally seen as likely to produce either the lowest or equal-lowest game total, and all but MoSSBODS have the Adelaide v GWS game as likely to produce the highest or equal-highest total. MoSSBODS prefers the Sydney v Essendon game for high total.

For low-scoring team we have West Coast according to MoSSBODS, GWS according to MoSHBODS, and Essendon according to the TAB and Centrebet bookmakers. All but MoSHBODS have Sydney as the round's high-scoring team. MoSHBODS prefers Adelaide with an expected score of 107 points.

By far the biggest difference between MoSSBODS' and the bookmakers' totals forecast comes in the Sydney v Essendon game, which for reasons unknown to me was priced at $1.90/$1.90 while the three other markets were priced at the usual $1.87/$1.87.

So, Investors have just the one unders/overs bet this week - an overs bet at $1.90 on the Swans v Dons game to produce at least 171 points.

In the three other games, the overlays are all 6 points or less. So, no bet.

The relatively subdued wagering in the overs/unders market is no bad thing this week if recent form is anything to go by, with MoSSBODS returning another worse-than-chance performance last week. It was on the right side of the total only twice with the TAB and just three times with Centrebet. That moved it to a 102 from 198 (52%) season-long record against the TAB and a 101 from 198 (51%) season-long record against Centrebet.

MoSHBODS fared even worse, landing just 2 from 9 against both the TAB and Centrebet to move to 101 from 198 (51%) season-long records against both.

In terms of mean absolute errors (MAEs), last week, MoSSBODS turned in the round's best mean absolute error (MAE) on game Margins (33.8 points per game) and on Away team scores (16.4 points per game), while Centrebet had the lowest MAEs for Home team scores (18.8), and Totals (13.2).

That gave Centrebet now the best season-long performance on game Margin (28.6) and still the best MAE for Home team scores (17.7), and left the TAB 1st on Away team scores (17.3) and on Totals (20.6).