Elo Predicts

This weekend there was only 5 NBL games on, so plenty of time to make some big updates to the site! NBL The NBL season simulation received the same treatment as the A-League and AFL. The prediction model accurately predicted 4/5 games in round 6. The team with the biggest gain was Melbourne United with +15 rating points. The New Zealand Breakers had the biggest rating point loss (-18) this round. A-League I have changed the way the season is predicted so season simulations take into account change in Elo rating. This means a more accurate (hopefully) prediction of how the season will play out. The A-League ladder also now includes the weekly change in Elo rating. Elo predictions shown in the season view are still based on a team's current rating, and I'm planning to leave it that way. AFL Likewise the AFL season simulation has been updated to include changes in Elo rating when simulating the season. Similar to the A-League, the percentage chance shown in the season view are ba

In a round without any draws, Elo predicted 5/5 in round 3. The big question is, can anyone catch Sydney (1661)? With a gain of 12 points this week they are 120 rating points ahead of 2nd-ranked Melbourne Victory (1541) and nearly 150 points ahead of 3rd-ranked Melbourne City (1516). The season simulation now has them a 100% chance to finish in the top 2, so something fairly catastrophic would have to happen to knock them out of the finals, even at this early stage of the season. Maybe too many draws? Home rating Home team Away Team Away rating Tip 1516 (-12) Melbourne City Sydney FC 1661 (+12) ✔ 1410 (-15) Wellington Phoenix Western Sydney Wanderers 1483 (+15) ✔ 1483 (-9) Newcastle Jets Melbourne Victory 1541 (+9) ✔ 1497 (+10) Perth Glory Brisbane Roar 1468 (-10)

Another round with more draws than outright winners, and I'm starting to think that I need to work on predicting the chance of a draw too. The Poisson distribution seems to do a reasonable job of that, but needs some work. Currently for the purposes of this blog I am counting a draw as a correct tip, however predicting a draw as a third option would be better I think. Since the 2011-12 season, when finals could no longer be drawn games, Elo ratings have an incredible 78% accuracy in finals so maybe just wait until the finals? Also worth noting, I heard one commentator say that Melbourne Victory were one of the favourites to win the Grand Final this season. To me it seems that Sydney is the clear standout at this stage of the season. Home rating Home team Away Team Away rating Tip 1495 (0) Adelaide United Newcastle Jets 1492 (0) ✔ 1391 (+4) Central Coast Mariners Melbourne City 1528 (-4)

There are usually winners and losers in the A-League, except this weekend which mostly ended in draws. In the case of a draw, it is the favourite who loses rating points. In round 1 the biggest loser was the Newcastle Jets who lost 13 rating point by going down to Wellington Phoenix. This loss sees Newcastle drop under the league average of 1500. Also interesting to note that Western Sydney had only a 50.1% chance to win, so the draw sees neither team gain or lose any rating points.  Home rating Home team Away Team Away rating Tip 1495 (+4) Adelaide United Sydney FC 1641 (-4) ✔ 1544 (-11) Melbourne Victory Melbourne City 1531 (+11) ❌ 1424 (+13) Wellington Phoenix Newcastle Jets 1492 (-13) ❌ 1480 (-3) Brisbane Roar Central Coast Mariners 1387 (+3) ✔ 1476 (0) Perth Glory Western Sydney Wanderers 1476 (0) ✔

Congratulations to the West Coast Eagles on their thrilling victory over Collingwood to close out the 2018 Season! Starting this Elo rating project has provided me with a new insight into team performance, and I hope the same for the readers of this blog. The rating system predicted 4 out of 9 finals, including correctly predicting the winner of the Grand Final. This is a lot lower than I would like, so there is room for improvement in the offseason. Like the BBL ratings I found that home ground advantage doesn't necessarily apply in the AFL, especially as West Coast were technically the home team (as they finished higher up the ladder) on Saturday but the game was played at Collingwood's home ground. Another change may be to move to an adjusted rating system similar to the BBL, where the rating is adjusted before the game to take into account the ground, rest days and other conditions. This would help I think in situations like the West Coast v Melbourne game where Melbourne