This week looks very similar to last week, with the MoS twins, on average, forecasting slightly lower totals than the bookmakers, though still differing by non-trivial amounts in a number of games.

The average expected margin for Round 18 is just 16.3 points per game, the lowest for any round since Round 14, and the lowest for any nine-game round since Round 10. It’s also the second-lowest average for a Round 18 across the period we’re analysing, and nine points below the all-season average for Round 18s across that same period.

MoSSBODS and MoSHBODS both elevated Brisbane Lions this week, MoSSBODS by one spot into 1st, and MoSHBODS by four spots into 4th.

MoSSBODS, in fact, shuffled a large number of teams in its top half, which left GWS in 2nd, behind the Lions, Richmond in 3rd, and Geelong in 4th. MoSHBODS, in contrast, left its Top 3 unchanged as GWS, Geelong and then Richmond.

In a round that, if anything, made it less certain which team is currently the best in the competition, just six of nine favourites won.

There’s a little more disagreement between MoSHPlay and MoSHBODS this week, but not much, with the margin forecasts differing only by amounts ranging from about 1 to 10 points, the largest of them being:

• Geelong v St Kilda: MoSHPlay Cats by 50; MoSHBODS Cats by 40

• Gold Coast v Adelaide: MoSHPlay Crows by 46; MoSHBODS Crows by 37

• Port Adelaide v Brisbane Lions: MoSHPlay Port by 26; MoSHBODS Port by 19

In only one game are MoSHPlay and MoSHBODS tipping different winners - the Essendon v North Melbourne game, which MoSHPlay has Essendon winning by 2, and MoSHBODS has North Melbourne winning by 4.

In something of a return to a milder version of the early parts of the season this week, the MoS twins are forecasting generally lower scores than the bookmakers, but overall only by a few points.

For the third week in a row the average bookmaker handicap for the round is above the season average, but it is, at least, under 20 points per game for the first time since Round 14, and for the first nine-game round since Round 10.

GWS, despite losing to Brisbane Lions, held onto 1st place on MoSSBODS and MoSHBODS this week, though it now has only a 1.1 Scoring Shot lead over the Lions on MoSSBODS, and a 4.6 points lead over the Cats on MoSHBODS.

MoSSBODS still has North Melbourne in 3rd, while MoSHBODS has this week rocketed Richmond into that position.

All of the Tipsters at the top of the Head-to-Head Tipster Leaderboard tipped 5 from 9 winners this week, which left MoSHPlay_Marg as leader, now on 90 from 135 (67%), still two tips clear of Bookie Knows Best and the RSMP twins.

MoSHPlay is, again this week, in close agreement with MoSHBODS about the likely margins in each of the nine contests, differing by a goal or more in only two of them:

• Carlton v Melbourne: MoSHPlay Melbourne by 14; MoSHBODS Melbourne by 6

• Western Bulldogs v Geelong: MoSHPlay Geelong by 19; MoSHBODS Geelong by 13

In all nine contests, MoSHPlay and MoSHBODS are currently selecting the same team to win, although MoSHPlay has Adelaide winning by only one point in their clash with Port Adelaide.

It’s another week of broad, overall agreement between the MoS twins and the bookmakers, though the twins have tended towards very slightly lower total scores.

Back to nine games for this week’s round, which always seems a much larger than 50% increase when you’re doing the data input.

No game this week has a bookmaker handicap of less than two goals, and only one has a handicap of much more than four, which has made for an all-game average of just over 20 points per game.

GWS, despite losing to Essendon this week, generated 21 scoring shots to the Dons’ 17, which was just enough to see MoSSBODS give them a pass mark and leave their Combined Rating essentially unchanged. MoSHBODS was less forgiving, however, and did drop the Giants’ Rating, but not enough to tip them out of 1st place.

Both Systems still have the same Top 2 then of GWS and Geelong, but MoSSBODS has the Roos in 3rd while MoSHBODS has retained Collingwood in that position.

There were no dramatic moves on the MoS Leaderboards this week, as the Head-to-Head Tipsters were separated by just one tip, the Margin Predictors by just 1.7 points per game, and the Head-to-Head Probability Predictors turned in very similar, slightly negative log probability scores.

MoSHPlay hasn’t fallen far from the tree with its forecast of a Giants victory by 14 points over the Dons, which is only 1 point different from its mentor’s, MoSHBODS, forecast.

This week, all nine totals markets were up when I checked at 2pm. Different week, different process, I guess.

The MoS twins have very similar views about the likely average total score across those nine contests, their average total coming within a point of the bookmakers’, but there are a number of individual games where the differences are starker.

Asking around, the politest explanation for this absence is forgetfulness, but were I running a multi-billion dollar enterprise I reckon I’d not accept that, as a boss, as a reasonable excuse. Those of you who’ve followed me for a while I hope might recognise that I’m, if anything, a forgiving soul. but this is, frankly, more than a bit embarrassing.

Anyway, on the assumption that this is correct, we have the MoS twins and the two bookmakers in complete agreement in terms of average totals.

Back to nine games for this week’s round, which always seems a much larger than 50% increase when you’re doing the data input.

No game this week has a bookmaker handicap of less than two goals, and only one has a handicap of much more than four, which has made for an all-game average of just over 20 points per game.

Geelong and Collingwood both saw their Combined Ratings decline this week on both Systems, which allowed GWS to extend its lead to 2.4 Scoring Shots over Geelong on MoSSBODS, and 6.8 points over Geelong on MoSHBODS.

Both Systems now have the same Top 3 of GWS, Geelong, and Collingwood, though, after that, the rankings become a lot more contentious.

In any other year, an all-Predictor average mean absolute error (MAE) of 21.5 points per game would be exceptional but, given the crazy-low MAEs we’ve seen earlier in the season, it’s hard not to feel just a little bit disappointed.

As I write this, the TAB is still to post a Totals market on the Saints v Lions game (and a bunch of other markets, besides), so for now I’ve assumed they’ll eventually put up a total close to, or the same as, Easybet’s. That’s certainly a reasonable extrapolation from what we’ve seen so far this round, but it is slightly weird that there’s no actual market from them.

Asking around, the politest explanation for this absence is forgetfulness, but were I running a multi-billion dollar enterprise I reckon I’d not accept that, as a boss, as a reasonable excuse. Those of you who’ve followed me for a while I hope might recognise that I’m, if anything, a forgiving soul. but this is, frankly, more than a bit embarrassing.

Anyway, on the assumption that this is correct, we have the MoS twins and the two bookmakers in complete agreement in terms of average totals.