Posts

Town 1-2 Derby

My Summary A good performance in both directions marred by a mistake and some poor finishing. While it was influenced by Derby not needing to score for 65 minutes 1.94-0.57 is still a very good xG scoreline and really quite unlucky to get the result we did. Attack I’ve seen a lot of suggestions that last night we were poor going forward, or at the very least no better than we used to be; so let’s take a closer look at this claim. Last night we created 1.94 xG,which for a start suggests we were quite unlucky not to score a second; on balance either Kachunga or Schindler should have scored. According to xG, last season we only produced more chances to score on one occasion, away at Bournemouth (an oddly similar game, conceding two silly goals early on, then getting one back quickly but not being able to finish the job). This would suggest that this attacking display was improved from last season, but this result alone is nowhere near definitive. For a start we got a penalty, ...
Post a Comment
Read more

A Plan B

Image
Recently I've seen more and more suggestions that it is time to move Mounie on, due to not fitting Seiwart’s system ; in my opinion that would be a mistake. It would leave Town without a viable plan B and take the club backward, to the same problem we had before Grant signed. So I want to look at Town's strikers and their style of play, to show where the similarities and differences lie. Quite often, for at least the last 2 seasons, Huddersfield have been criticised for lacking a plan B up front. During the majority of that time, the only 2 players used as strikers were Mounie and Depoitre, who have extremely similar styles of play, so much so that they seemed to cancel each other out when playing together. There’s clear evidence to show this:  I ran a clustering analysis on the attacking and build up data from 230 strikers in Europe's top 5 leagues to see just how similar they were. The results are below. Although the names aren’t very clear, it is important ...
Post a Comment
Read more

Premier League Predictions 26

So now using version 3 Week 26 Predictions - Home Predicted Score Away Fulham 1-2 Man United Crystal Palace 1-0 West Ham Huddersfield 0-1 Arsenal Liverpool 2-0 Bournemouth Southampton 1-0 Cardiff Watford 2-1 Everton Brighton 1-0 Burnley Tottenham 1-0 Leicester Man City 1-0 Chelsea Wolves 1-0 Newcastle Still a lack of draws - I think I may need to change the data or at least the form it is in to help it predict more accurately. Not many surprising results here with most teams you would expect to win being predicted to and lots of tight games for the rest. Week 25  Review - Home Predicted Score Actual Score Away Tottenham 1-0 1-0 Newcastle Brighton 2-1 0-0 Watford Burnley 2-1 1-1 Southampton Chelsea 1-0 5-0 Huddersfield Crystal Palace 1-0 2-0 Fulham Everton 2-1 1-3 Wolves Cardiff 0-1 2-0 Bournemouth Leicester 1-0 0-1 Man United Man City 1-0 3-1 Arsenal West Ham 1-2 1-1 Liverpool Everton 0-1 0-2 Man City Another good week with 5 results and 1 corre...
Post a Comment
Read more

Confidence - an xG viewpoint

Image
In the early months of the season Huddersfield were massively under-performing and the xg stats suggested that their form would increase dramatically and it did with 7 points taken in November but since then both actual and expected performance have dropped significantly and I think a possible explanation is confidence - that being below our xg totals for so long has actually made the squad believe in themselves less and thus produce less quality chances so our xg output sunk and thus are actual goals scored stayed low instead of rising up to the xg average. So I wanted to look and see if it is a common occurrence for clubs xg to move to meet ag rather than the other way round which is to the best of my knowledge what you would expect for a team under or over performing their expected goal scoring ability Take a look at 5 game rolling expected goals (xg) and actual goals (ag) for both Southampton and Huddersfield this season: Southampton as - 5 game rolling average actual goa...
Post a Comment
Read more

Premier League Predictions 25

So now using version 3 Week 25 Predictions - Home Predicted Score Away Tottenham 1-0 Newcastle Brighton 2-1 Watford Burnley 2-1 Southampton Chelsea 1-0 Huddersfield Crystal Palace 1-0 Fulham Everton 2-1 Wolves Cardiff 0-1 Bournemouth Leicester 1-0 Man United Man City 1-0 Arsenal West Ham 1-2 Liverpool Everton 0-1 Man City Still no draws not sure why the behaviour is so weird around this - works perfectly fine for other seasons. Chelsea perhaps surpisingly only backed to beat Huddersfield by a small margin but mostly due to their lack of goals in recent weeks. Leicester's impressive performances against big six clubs recently mean they have been backed again against Man United. Week 24  Review - Home Predicted Score Actual Score Away Arsenal 2-0 2-1 Cardiff Fulham 1-0 4-2 Brighton Huddersfield 1-0 0-1 Everton Wolves 2-1 3-0 West Ham Man United 1-2 2-2 Burnley Newcastle 0-1 2-1 Man City Bournemouth 1-0 4-0 Chelsea Southampton 1-0 1-1 Crystal Palace Li...
Post a Comment
Read more

Premier League Predictions 24

So now using version 3 a summary of its performances over recent weeks and its predictions for the upcoming games Week 17-23  Review - Week 17 - 7 results with 2 scores Week 18 -  3 results with 0 scores Week 19 -  6 results with 1 score Week 20 -  5 results  with 2 scores Week 21 -  3 results with 0 scores Week 22 -  6 results with 2 score s Week 23 -  7 results 0 scores Overall a fairly good set of results with just a few poor weeks in the middle. Scores very low with this model but getting a clear improvement in results. Running just above the average for last season with 37/70 results. Week 23 Predictions - Home Predicted Score Away Arsenal 2-0 Cardiff Fulham 1-0 Brighton Huddersfield 1-0 Everton Wolves 2-1 West Ham Man United 1-2 Burnley Newcastle 0-1 Man City Bournemouth 1-0 Chelsea Southampton 1-0 Crystal Palace Liverpool 2-0 Leicester Tottenham 2-1 Watford A huge surprise there with Burnley to win away at Man ...
Post a Comment
Read more

Predictor Version 3

Version 3 The new predictive element I have calculated to be the best to use is a logistic regression machine learning element to predict result of a match. Followed  by taking the average xG scored and conceded by both the home and away team to produce an average value for the home team to score and away team to score which then uses a Poisson distribution with these numbers as mean to find the most likely score line with matching result to the first element. Naive Bayes and Results was selected by testing performance against 3 other predictors that use either Bayes or Logistic regression to predict either scoreline or result. The performance test was accuracy of predicting the correct result even if scoreline was used to train the data as ultimately that is what they are both trying to predict. Results shown below: ML\Predict Results Score Bayes 0.47 0.31 Logistic 0.50 0.47 So this is only a small improvement from the old version 2 predictor which got 0.46...
Post a Comment
Read more