| 1st half of the season: An analysis 15:30 - Dec 31 with 2538 views | quantfootball |
Hi all,
I had done some analysis on the first 23 games of the season, looking at our underlying metrics and it ended up becoming an interesting piece. I've been given permission to post a link to the substack I uploaded it to:
https://footballquant.substack.com/p/the-ipswich-paradox
thanks |  | | |  |
| 1st half of the season: An analysis on 15:54 - Dec 31 with 2282 views | Ewan_Oozami |
Thank you - I think that very succinctly reflects my thoughts on the season so far and what I expect to happen for the rest of the season! |  |
| |
| 1st half of the season: An analysis on 15:57 - Dec 31 with 2274 views | vilanovablue |
A very enjoyable analysis, good stuff. |  | | |
| 1st half of the season: An analysis on 16:54 - Dec 31 with 2157 views | ringwoodblue |
A very interesting, insightful, well-explained and illustrated analysis. Thanks for sharing. |  |
| |
| 1st half of the season: An analysis on 17:14 - Dec 31 with 2086 views | PhilsAngels |
mind blown | | | |
| 1st half of the season: An analysis on 17:50 - Dec 31 with 2013 views | mellowblue |
Thank you, very interesting. Football being so analysis-driven now, you can see why there is no place in football for dinosaurs like Allardyce, McCarthy etc. The profile of managers has changed. | | | |
| 1st half of the season: An analysis on 18:57 - Dec 31 with 1938 views | djgooder |
In my Uber summary it says we are awesome, but also we haven’t performed to the levels we should have so shouldn’t feel like we are awesome.
Kind of how I feel I guess but could never have described it in such a way.
A really good read. Thanks. |  | | |
| 1st half of the season: An analysis on 19:36 - Dec 31 with 1868 views | KungfuBlue |
Enjoyed the analysis. Thanks for the effort you put in and for sharing |  | | |
| 1st half of the season: An analysis on 19:43 - Dec 31 with 1850 views | MK1 |
Brilliant work. Thanks. |  |
| |
Login to get fewer ads
| 1st half of the season: An analysis on 20:37 - Dec 31 with 1771 views | NthQldITFC |
That is really, really good, thanks. Thorough and well written. Nice job. |  |
| |
| 1st half of the season: An analysis on 23:44 - Dec 31 with 1637 views | Kropotkin123 |
Awesome, thank you. Great read. |  |
| Submit your 1-24 league prediction here -https://www.twtd.co.uk/forum/514096/page:1 - for the opportunity to get a free Ipswich top. | | Poll: | Would you rather | | Blog: | Round Four: Eagle |
| |
| 1st half of the season: An analysis on 06:19 - Jan 1 with 1498 views | Battersea_Blue |
That's really interesting stuff, thanks for sharing. | | | |
| 1st half of the season: An analysis on 09:20 - Jan 1 with 1370 views | Dubtractor |
That's a really well written analysis, and does a good job of explaining why xg etc isn't bollards, even if some don't like it as a metric.
A lot of what you write is increasingly apparent from the visual test, in particular the low number of good chances that we give away.
Just need to be more clinical with the chances that we do create! |  |
| |
| 1st half of the season: An analysis on 10:42 - Jan 1 with 1247 views | Alberto_the_frog |
Too clever for me. I think it agrees with what I thought just looking at results and highlights, though. Southampton and Sheff U would be expected to have more points. I'd put Leicester in that category too. |  | | |
| 1st half of the season: An analysis on 11:03 - Jan 1 with 1195 views | NthQldITFC |
| 1st half of the season: An analysis on 10:42 - Jan 1 by Alberto_the_frog |
Too clever for me. I think it agrees with what I thought just looking at results and highlights, though. Southampton and Sheff U would be expected to have more points. I'd put Leicester in that category too. |
The level of agreement with the eye test to some extent helps to validate the methodology, as does correlation with completed seasons, as long-term as possible, over several seasons of data gathering.
Then you have confidence that the methodology has value, not as a precise predictor, but as better and better tool to help understand what's likely to happen. And if it works well for your own club, for which you inevitably have far more visual experience and ability to judge, then you can have a high degree of trust that it really helps understanding other clubs of which you have limited visual experience. | |
| |
| 1st half of the season: An analysis on 12:27 - Jan 1 with 1106 views | Smoresy |
Top, top stuff.
My only (personal) caution would be on regression to the mean within a single season. Hull enjoy the greatest positive variance currently in points compared with "expected points", +11 now I believe. In 24/25, the highest positive variances were welcomed by Burnley, +24 points, and Sheff Utd, +20. In 23/24, we were the biggest "beneficiaries" with +15. In 22/23, it was Burnley again with +19. 21/22, Huddersfield with +22, lovely.
So hopefully variance evens out - good reason to hope it will - but it's also normal to finish a season with notable outliers in both directions, either through luck, individual excellence or a combination. None of this negates the positivity which rightfully flows through your piece. The numbers are with us! (All hail the numbers.) | | | |
| 1st half of the season: An analysis on 12:34 - Jan 1 with 1081 views | NthQldITFC |
| 1st half of the season: An analysis on 12:27 - Jan 1 by Smoresy |
Top, top stuff.
My only (personal) caution would be on regression to the mean within a single season. Hull enjoy the greatest positive variance currently in points compared with "expected points", +11 now I believe. In 24/25, the highest positive variances were welcomed by Burnley, +24 points, and Sheff Utd, +20. In 23/24, we were the biggest "beneficiaries" with +15. In 22/23, it was Burnley again with +19. 21/22, Huddersfield with +22, lovely.
So hopefully variance evens out - good reason to hope it will - but it's also normal to finish a season with notable outliers in both directions, either through luck, individual excellence or a combination. None of this negates the positivity which rightfully flows through your piece. The numbers are with us! (All hail the numbers.) |
That's interesting. I wonder if that points to a tweak which ought to be made to tune up the xG base of the calculation? | |
| |
| 1st half of the season: An analysis on 13:12 - Jan 1 with 1022 views | DarkBrandon |
Excellent stuff.
Rarely has my scepticism at clicking on a link to a random “here is my blog” link been so thoroughly misplaced.
As others have said though variance can last a looong time. I wouldn’t be surprised if you could have three years of good luck and still not be wildly exceptional. |  | | |
| 1st half of the season: An analysis on 13:26 - Jan 1 with 975 views | backwaywhen |
| 1st half of the season: An analysis on 17:14 - Dec 31 by PhilsAngels |
mind blown |
Headaches bad . | | | |
| 1st half of the season: An analysis on 13:31 - Jan 1 with 964 views | Smoresy |
| 1st half of the season: An analysis on 12:34 - Jan 1 by NthQldITFC |
That's interesting. I wonder if that points to a tweak which ought to be made to tune up the xG base of the calculation? |
Difficult because the attempt is to accurately calculate goal-scoring probabilities for the average shot-taker against the average goalkeeper. In any league there will be players who consistently perform above the average in these key departments, others who experience career highs or lows, etc.. Stoke's Johansson is a good example of a goalkeeper who outperforms his xGOT calculation every season, and does so emphatically more often than not.
Furthermore, xG doesn't account for the order in which goals are scored. We know how important the first goal is for a typical game's outcome, but also for how often it impacts on play thereafter. Take Hull v Boro a few days ago: Hull scored early and elected to defend for over an hour. It isn't the business of xG to reflect situational tactics of course, but it remains relevant and unaccounted for. Burnley 24/25 is likely a good example of an xP outlier due to tactical behaviours: score first then death by boredom to the opponent. | | | |
| 1st half of the season: An analysis on 13:52 - Jan 1 with 916 views | Plums |
Thank you, that's an excellent and informative read.
Really adds something to the 'where are we' debate that's ongoing atm. |  |
| |
| 1st half of the season: An analysis on 14:40 - Jan 1 with 847 views | Pinewoodblue |
Thanks for the time you must have spent in this analysis.
It also supports my gut feeling that neither Coventry, nor B’oro will avoid the play offs. |  |
| |
| 1st half of the season: An analysis on 10:31 - Jan 2 with 701 views | DJR |
One thing that I believe xG fails to take into account is the likelihood of the person in question making the most of a chance.
For example, Harry Kane is more likely to score a 0.5 chance than many other strikers.
In our case, I do wonder if part of the problem is that quite a few of our players don't seem to have that killer instinct when in a possible goal-scoring position. That is clearly not true of Clarke or Philogene but there are others who seem rather goal shy.
By way of example, Wes Burn put away the chance against Coventry in a way that might be expected of him but I could imagine that if that chance had fallen to certain others, they might not have even hit the target. [Post edited 2 Jan 10:32]
|  | | |
| 1st half of the season: An analysis on 11:27 - Jan 2 with 657 views | Steve_M |
Interesting analysis, thanks. |  |
| |
| 1st half of the season: An analysis on 17:45 - Jan 7 with 365 views | quantfootball |
Thank you all for the kind comments! Unfortunately, I have caught a new year illness, but the next part should be complete in a couple of weeks. There are also some very insightful comments regarding xg, and regression, that I have had some thoughts on and would like to dig into further! | | | |
| 1st half of the season: An analysis on 18:51 - Jan 7 with 268 views | thebooks |
| 1st half of the season: An analysis on 17:50 - Dec 31 by mellowblue |
Thank you, very interesting. Football being so analysis-driven now, you can see why there is no place in football for dinosaurs like Allardyce, McCarthy etc. The profile of managers has changed. |
Allardyce was one of the most inventive, advanced managers of his time.
As for the OP, it was really good, got me thinking. I’m not entirely sure those stats quite match what I’m seeing, in as much as we do clearly follow a process, generate chances and then fade away in games and struggle coming up with something when the process doesn’t work. It’s like we invite a couple of decent chances through this kind of fear.
Still, I do agree that over the whole season, Cov, Boro and Hull will fade away, and we’ll end up winning the league quite comfortably. |  | | |
| |
By continuing to use the site, you agree to our use of cookies and to abide by our Terms and Conditions.
