Serie A Predictions Week 33

Serie A 429

GameHome Team Win %Draw %Visiting Team Win %Actual OutcomeCorrect Prediction
Udinese v. Inter Milan31%41%27%Udinese 1 - Inter 2No (3rd most likely)
Milan v. Genoa17%56%27%Milan 1 - Genoa 3No (2nd most likely)
Juventus v. Fiorentina49%36%14%Juve 3 - Fiorentina 2Yes
Chievo v. Cagliari77%16%7%Chievo 1 - Cagliari 0Yes
Cesena v. Atalanta20%48%32%Cesena 2 - Atalanta 2Yes
Palermo v. Torino12%37%51%Palermo 2 - Torino 2No (2nd most likely)
Sampdoria v. Verona48%36%16%Sampdoria 1 - Verona 1No (2nd most likely)
Lazio v. Parma76%18%6%Lazio 4 - Parma 0Yes
Sassuolo v. Roma8%28%64%Sassuolo 0 - Roma 3Yes
Empoli v. Napoli5%56%39%Empoli 4 - Napoli 2No (3rd most likely)


La Liga Mid-Week Predictions

GameHome Team Win %Draw %Visiting Team Win %Actual OutcomeCorrect Prediction
Barcelona v. Getafe87%9%4%Barcelona 6 - Getafe 0Yes
Athletic v. Real Sociedad63%21%15%Athletic 1 - Real Sociedad 1No (3rd most likely)
Levante v. Cordoba45%34%21%Levante 1 - Cordoba 0Yes
Eibar v. Sevilla42%22%35%Eibar 1 - Sevilla 3No (2nd most likely)
Celta Vigo v. Malaga19%35%45%Celta Vigo 1 - Malaga 0 No (3rd most likely)
Real Madrid v. Almeria93%4%4%Madrid 3 - Almeria 0Yes
Elche v. Deportivo57%29%13%Elche 4 - Deportivo 0Yes
Villareal v. Atletico Madrid20%61%19%Villareal 0 - Atletico 1No (3rd most likely)
Rayo v. Valencia25%8%66%Rayo 1 - Valencia 1No (3rd most likely)
Granada v. Espanyol13%37%48%Granada 1 - Espanyol 2Yes

Predictions of La Liga Winner: Pre-Week 33

I ran the simulations for the run-in for each of the top 3 teams in Spain right now, and Barcelona’s 2 point lead over Real Madrid translates into a 68% chance of finishing the season on top of La Liga. Madrid has a 17% chance of winning the league outright, and an 11% chance of finishing the season with the same number of points as Barcelona. 1

La Liga Week 33

Other interesting stats: Barcelona has about a 33% chance of winning the rest of their games and finishing with the max 93 points, while Madrid only has a 26% chance of winning out and finishing with 91. Atletico Madrid still has about a 3% chance of winning the league

  1. La Liga’s tiebreaker is different than the typical “goal differential”, where they look at results between the tied teams first, and Madrid has a +1 goal differential in the two games

La Liga Predictions – Week 32

I’m a little late posting the predictions for this week because I was putting together the probabilities for the rest of the season’s fixtures, but so far so good. I’ve got 4 correct predictions, 1 that was the 2nd most likely outcome (but only off by 3% so really it’s a push).

All the games today were relatively easy, although it predicted a solid win by Athletic of Getafe so I’m happy with that one. The rest of the predictions are in the table below:

GameHome Team Win %Draw %Visiting Team Win %Actual OutcomeCorrect Prediction
Levante v. Espanyol38%35%26%Levante 2 - Espanyol 2No (2nd most likely)
Barcelona v. Valencia70%14%15%Barcelona 2 - Valencia 0Yes
Real Madrid v. Malaga84%5%10%Real Madrid 3 - Malaga 1Yes
Deportivo v. Atletico Madrid6%26%67%Deportivo 1 - Atletico Madrid 2Yes
Athletic Bilbao v. Getafe CF76%18%6%Athletic 4 - Getafe 0Yes
Rayo v. Almeria81%6%13%Rayo 2- Almeria 1Yes
Granada v. Sevilla6%31%63%Granada 1 - Sevilla 1No (2nd most likely)
Villareal v. Cordoba59%40%2%Villareal 0 - Cordoba 0No (2nd most likely)
Eibar v. Celta Vigo56%21%21%Eibar 0 - Celta Vigo 1No (2nd most likely)
Elche v. Real Sociedad42%32%24%Elche 1 - Real SociedadYes

EPL: Week 33 Predictions

Here are my Week 33 predictions.  The big game this weekend is Chelsea v. Manchester United. The model actually has what seems like a reasonable prediction here, with Chelsea a 46% favorite at home, 27% to draw and 27% for Manchester United.

On the other end of the table, all of the relegation teams have a reasonable chance of at least getting a point, while Burnley has the strongest predicted chance of getting the three points at Everton. Everton has come on strong lately, and they’re a 45% favorite, but I wouldn’t be surprised if Burnley got something out of this match at all. This could be a key week in determining the relegation spots, as with all the teams so close even a point could be huge.

GameHome Team Win %Draw %Visiting Team Win %Actual OutcomeCorrect Prediction
Stoke City v. Southampton25%26%49%Stoke City 2 - Southampton 1No (3rd most likely)
Leicester City v. Swansea City12%14%73%Leicester City 2 - Swansea 0 No (3rd most likely)
Everton v. Burnley45%24%30%Everton 1 - Burnley 0Yes
Crystal Palace v. West Brom31%36%32%Crystal Palace 0 - West Brom 2No (2nd most likely)*
Chelsea v. Manchester United46%27%27%Chelsea 1 - Manchester United 0Yes
Manchester City v. West Ham56%29%14%Manchester City 2 - West Ham 0Yes
Newcastle v. Tottenham14%26%59%Newcastle 1 - Tottenham 3Yes
Hull City v. Liverpool6%35%58%
Arsenal v. Sunderland62%18%19%Postponed until May 20 -

Will Manchester City Keep their Champions League Place?

With the news that Manuel Pellegrini will be fired if Manchester City fails to earn a Champions League place this season (if not sooner), I wanted to look and see what the odds were of that actually happening.

The Method: I calculated the predicted probabilities for each of the top 8 teams’ remaining games, and then simulated 10,000 seasons based on those probabilities.  I added the predicted points for each game to the teams’ current totals, then ranked the teams based on those totals.  Finally, I calculated the proportion of times each team ended in each position 1, and then added up the probability of finishing in each of the top 4 spots to calculate the likelihood of finishing in the Champions League positions. 2.

As of today, my simulations are showing Manchester City has an 86% of keeping a top 4 spot (assuming they win the tiebreaker). However, if they fail to defeat West Ham where they are a 56% favorite, their likelihood drops to 73%. And more troubling, if they lose on the road to a desperate QPR team where they are a 62% favorite, their likelihood drops below 50%.

Man City Top $

The good news is that if Manchester City can win the next two games, they’re pretty much guaranteed a Top 4 spots. The results of the simulation are here:

Man City Top 4 Good

However, City fans shouldn’t panic if they lose the next game against West Ham. If they win a very winnable game against QPR week 34 they’ll have an over 80% chance to keep a Champions League spot.

Man City Top 4 Middle

All of this depends on other results, but the next two games are crucial for Manchester City. They’ll need to shake off this bad form, especially because the next two games are against relative minnows who Champions League caliber teams should have no problems defeating. Two big weeks coming up for Pellegrini if he wants to have any hope of surviving through the season.

  1. Good news Chelsea fans, you have a 98% chance of winning the league at this point
  2. Because my simulator doesn’t calculate final scores, I counted a tie for 4th place as a Champions League place for all the teams

Champions League Places – Week 33

This was a big week for movement in the Champions League spots: Arsenal’s win bumped them from a 71% chance to qualifying for UEFA’s top competition to a 91%, while Spurs were the biggest loser of the week, dropping from a 17% chance to only 7%.

Southampton had a good week as well, moving up 1% point but gaining about a 26% chance to end up with the same number of league points as Manchester City. City currently has a significantly higher goal differential, but if they lose enough games to let Southampton catch up they could also lose that tie-breaker so that has to be a concern for fans of the Citizens.

I have Liverpool as a decent favorite tomorrow, so it will be interesting to see what happens to their chances after tomorrow’s game. It has to be a must-win for Liverpool if they want to have any realistic chance of making the top 4, and home against Newcastle is certainly a winnable game. The plot includes probabilities before this weekend’s fixtures at the top, and probabilities after Sunday’s results at the bottom.

Top 4 Plot Week 32

La Liga Week 31 Predictions

I’ve finished up my preliminary spreadsheets for La Liga, so I thought I’d post the predictions here as well. So far the model is doing ok, although Barcelona’s draw at Malaga was a big miss.

I’ll admit to not knowing enough about Spanish soccer to comment much on the model’s predictions, but I did a quick check to see how things look at the top of the table and it seems to predict pretty well.  I’ll update the results as the weekend goes on, and I’m hoping to add Serie A and the Bundesliga soon. Serie A might be a little painful because I’m not sure I’m ready to relive Milan’s season just yet, even in spreadsheet form.

GameHome Team Win %Draw %Visiting Team Win %Actual OutcomeCorrect Prediction
Sevilla v. Barcelona38%8%54%Sevilla 2 - Barcelona 2No (3rd most likely)
Almeria v. Granada
46%33%20%Almeria 3 - Granada 0Yes
Celta Vigo v. Rayo Vallecano32%28%40%Celta Vigo 6 - Rayo 1No (2nd most likely)
Real Madrid v. Eibar91%4%5%Real Madrid 3 - Eibar 0Yes
Malaga v. Atletico Madrid29%14%57%Malaga 2 - Atletico 2No (3rd most likely)
Getafe CF v. Villareal35%15%50%Getafe 1 - Villareal 1No (3rd most likely)
Cordoba v. Elche50%32%17%Elche 2 - Cordoba 0No (3rd most likely)
Real Sociedad v. Deportivo La Coruna55%32%13%Real Sociedad 2 - Deportivo 2No (2nd most likely)
Espanyol v. Athletic Bilbao64%8%27%Espanyol 1 - Athletic 0Yes
Valencia v. Levante77%20%3%Valencia 3 - Levante 0Yes

Week 32 Predictions

The week 32 predictions are done, and I have a few thoughts.

First, after a so-so performance last week I revised the model a little bit to include both previous results and a few covariates (see my previous post on the topic for more details). This performs better in all the diagnostic tests, so hopefully I’ve added a little predictive accuracy to the model.

Second, there are two particularly…interesting predictions this week. The model is predicting Chelsea as almost a 100% favorite to win on the road against QPR this weekend. They’re obviously a huge favorite, but I honestly don’t know if Chelsea would be a 99% favorite against a mid-table team in the Championship, let alone a Premier League team fighting for its life. The other one is Manchester City being a favorite against Manchester United at Old Trafford. I think my model consistently underestimates Man United because they lost at least a few games they should have won early in the season. I tried a specification where I added recent form into the model, but it didn’t perform as well as the GF/GA variable so I took it out. If I had to guess, it probably helped predict Manchester United results a little more accuracy though. As a City fan, I have to think Man United is the favorite this weekend, especially since Man U seems to have figured things out. Vincent Kompany and (to a MUCH lesser extent) Wilfried Bony being out make it a tougher challenge, but the model doesn’t know that.

Nothing else looks too out of the norm, so we’ll see how it does this weekend!

GameHome Team Win %Draw %Visiting Team Win %Actual OutcomeCorrect Prediction
Swansea City v. Everton46%22%31%Swansea 1 - Everton 1No (3rd most likely)
West Ham v. Stoke City33%43%23%West Ham 1 - Stoke City 1Yes
Southampton v. Hull City79%6%15%Southampton 2 - Hull City 0 Yes
Burnley FC v. Arsenal6%34%62%Burnley 0 - Arsenal 1Yes
Sunderland v. Crystal Palace21%52%26%Sunderland 1 - Crystal Palace 4No (2nd most likely)
West Brom v. Leicester City35%44%20%West Brom 2 - Leicester City 3No (3rd most likely)
Tottenham v. Aston Villa69%14%16%Tottenham 0 - Aston Villa 1No (2nd more likely)
QPR v. Chelsea1%0%99%QPR 0 - Chelsea 1Yes
Man United v. Man City12%44%43%Man United 4 - Man City 2No (3rd most likely)
Liverpool v. Newcastle69%7%24%Liverpool 2 - Newcastle 0Yes

Week 31 Recap/Reassessment

So  the model didn’t perform as well Week 31 as it did Week 30. It actually did about as well as “chance” with 3 correct predictions, 3 times the second most likely outcome occurred, and 4 times the third most likely outcome occurred.

So what happened? First, the correct predictions: Chelsea unsurprisingly won (the model REALLY likes Chelsea at home against anyone). Swansea over Hull City wasn’t much of a surprise either, and while it wasn’t much of an upset, the model did well on the marquee game of the weekend: Arsenal v. Liverpool. That’s actually a pretty tough game to pick, although Liverpool seems to have returned to form after winning 12/13, and after a slow start Arsenal has been as good this year as they have been in 5 years.

Objectively, the 2nd most likely outcome picks were pretty solid. I think intuition would have had Tottenham as a much bigger favorite over Burnley, so predicted a 1/3 chance of a draw seems pretty solid. I’m actually really pleased with the QPR win over West Brom: I didn’t see the betting line, but I have to think QPR was a pretty significant underdog on the road against a decent West Bromwich Albion team. Giving them a 29% chance of winning, and a 45% chance of a draw is more than I would have done.

Manchester United represents a specific challenge for my model: I posted that it seemed off that they were only a 33% likelihood to win at home against Aston Villa, and my instinct was right there.  The model consistently underrates Manchester United, probably likely due to their uninspiring early season form. It’s doing so again this weekend at home against slumping Manchester City, but we’ll see what happens.

Finally, the complete misses. Leicester City over West Ham, and Crystal Palace over Man City have to be considered pretty big upsets (although Palace has looked good recently). I’m not sure there’s anything I can do about results like that, and the model does quantify the unlikely nature of these types of results.

Everton v. Southampton and Sunderland v. Newcastle have to be considered bad misses. Southampton’s been in good form this year and Everton is obviously underperforming, but giving them 8% to win at home has to be considered a miss. Sunderland v. Newcastle was a tough one, but giving Sunderland 13% to win at home seems a bit of an underestimation of their chances.

I’ve made some changes in the model this week: specifically I’ve added covariates to the prediction model. I played with a few different models that included recent form, transfer spending, and goals scored/goals allowed, and the best model in terms of predictive power and parsimony was simply predicting total points as a function of goals for and goals allowed. Transfer spending added a little predictive power, and recent form surprisingly only added a little predictive power (I was actually concerned it was too collinear with the outcome variable but that turned out not to be true), so for now I’m leaving it with just the two explanatory variables. This summer and fall will be devoted to using player data to predict outcomes, so those will likely go into the model as well at some point, but for now I’m happy with it. A quick diagnosis shows it gives Manchester United a little more credit (it gave them a 38% win likelihood last weekend and something like 42% chance to draw…still a 2nd best prediction, but closer), so I’m happy with it for now.

Looking forward to seeing this week’s outcomes and how well the model does. I like most of the picks, so we’ll see what happens.