Moneyball in Football Manager 2019: III

moneyball logo

Before we kick off with the 3rd update of this save, I just want to say thank you very much to everyone that has plugged, promoted, liked or read these updates so far. I’ve had a fantastic response to the save so far, and this is the most successful this blog has been – in terms of views – since my beloved Bayer Leverkusen save back on FM15. Anyway, back to Moneyball – in case you’ve missed this save so far, here are the first and second updates. In the last update, I covered the first half of the season, which had gone very well, with us five points clear of Celtic at the top of the league, into the Knockout Round of the Europa League against Hoffenheim. We had also made some classic Moneyball transfers, bringing in 4 complete unknowns, whilst moving out those who had statistically proven that they weren’t good enough to play for Rangers.

The conditions or rules that I’ve applied to my save come largely from the concepts outlined in Moneyball, but also from the excellent football economy books Soccernomics by Simon Kuper and Stefan Szymanski, and The Numbers Game by Chris Anderson and David Sally. If you need a reminder, below are the rules I’m using – this is the last time I will post these in the actual article, from now on they will be added as a link:

  1. Players are undervalued for a range of biased reasons such as age, personality, appearance. None of these matter. All that matters is production on the field, which can be analysed through statistics.
  2. Several countries are overvalued in the market, such as England, Holland and Brazil and there is better value for money off the beaten track.
  3. Wage spend correlates to success far more than transfer expense.
  4. Prioritise identifying and improving your weak spots in transfer windows. This is the best way to improve your team.
  5. Strikers cost more than they should and can be poor value for money.
  6. Ensure that all potential signings are thoroughly scouted – the wisdom of crowds – before their statistics can be analysed. 
  7. Don’t buy players who performed well solely at international competitions. These are a perfect example of ‘small sample size’, where stats have no true value to them as they can’t be trusted to be replicated. 
  8. Long term, producing your own players and developing through the academy is more cost effective. 
  9. Football is won by scoring more goals than the opposition – like baseball is won by scoring more runs – and therefore attacking stats must be a main focus.
  10. Some of the best value can be found in identifying players who are no longer wanted by their clubs, and are likely to vastly undervalue them.

Anyway, on with Rangers, and the end of our first season at Ibrox. As mentioned above, when we left off we were 5 points clear of Celtic – so how did this season end?

Screen Shot 2019-02-19 at 12.52.05Well, not only did we manage to hold onto our lead at the top, but we extended it to 11 points by the end, miles away from Celtic. We’ve absolutely dominated the league in the end, finishing with the best goal difference comfortably with +54, beginning to score more goals towards the end of the season to match our excellent possession, ending 2nd for goals scored, and 1st still for goals conceded. We remained top of the charts for possession as well, ending with an average possession of 62%. When these are combined, they match very well with my theory and statistical evidence of prioritising attacking and possession, although ideally we’d be 1st for goals scored and 2nd for conceded, although undoubtedly this system and setup has served us very well.

My only complaint about our league season is our performances against Celtic. Whilst we may technically be 11 points better than our Old Firm rivals, I don’t believe we’re necessarily better than they are yet. In the previous update I said that we still hadn’t beaten them, and that has continued to the end of the season. We’ve drawn the majority of the games against them this season – including the Old Firm Derby just after we clinched the league title – but next season I want to beat them at least once, and they’ll be back next season for sure.

Screen Shot 2019-02-19 at 12.58.30

Unfortunately, our Europa League run did come to an end at the hands of Hoffenheim in the First Knockout Round. We went down 1-0 away in the 1st leg, and despite an impressive performance in the second leg – and a screamer from Jordan Rossiter, who’s been able to play more in the second half of the season – only managed to draw 1-1 at Ibrox. This was a shame for sure, but it allowed us to focus entirely on the league, and hopefully we can make the step up to Champion’s League quality next season, or probably end up joining the Europa League at a later stage.

What I’m going to do now, is analyse the signings I’ve made, and look at whether my transfer dealings have worked, and whether Moneyball is working. I may have won the league, but that doesn’t necessarily mean that the concept is working, because as many that are involved in statistics – particularly in baseball – will tell you, it’s about the process, not the outcome. Firstly, lets look through my sales this season.

Screen Shot 2019-02-19 at 13.25.17

As you can see, I’ve made smart sales, selling players who statistics told me weren’t playing well, and those trends have continued at other teams. Candeias hasn’t performed very well at Ligue 1 level for Guincamp with a 6.62 average rating. Andy Halliday has done much the same for Hibs, only getting a 6.68 rating. Graham Dorrans performed slightly better for Hibs, but is still only rated at 6.86 overall, far below what I’m looking at as an acceptable level for my Rangers. Carlos Peña did average about 7.00 when he was sold to Dorados, but remember, they’re only a second division Mexican team and he has only made 4 appearances, so this is a perfect example of small sample size. Kyle Lafferty hasn’t bucked the trend at FC Zurich, averaging 6.78, and Allan McGregor proved to be an excellent sale, so far he’s only made one appearance for Spurs, and averaged a good 7.00 in that match. But still, that’s at least £1.3m for one game. Value for money wise, I’ll take that. Now, let’s analyse our purchases this season. I haven’t included those bought before I came to the club such as Connor Goldson, because they weren’t bought under the Moneyball principles.

Screen Shot 2019-02-19 at 13.36.48

All in all, things have gone pretty well here. My first signing Andreas Skov Olsen has averaged 7.26 across the season, scoring 13 goals and 5 assists in 33 league appearances this season. He’s now valued at £1.8m. Lucas Piazón has recovered after a slow start to average 7.08 over the season with 7 goals and 6 assists, and his value has now jumped up to £4.4m and if anyone comes in for him, I’ll happily cash in at this point. Óscar Gil started off very well but then tailed off towards the end of the season but still ended with a 7.03 rating, with 92% pass completion and 88% tackle success. He’s now worth £3.4m. However, I’m still looking at the defensive midfielder role as an area of weakness moving forward. Alex Fernandez wasn’t available to play all season because of his cruciate ligament injury, but I look forward to getting him on the field next season, because he now faces serious competition for his spot that I was confident would be assured. Lawrence Shankland only averaged 6.86, but was brought on almost entirely as a sub, and therefore didn’t have much time to contribute, but did score 4 goals in 9 games, so despite the lower average rating, he produced when I asked him to. Again though, like defensive midfield, I’m now looking at striker as a weakness – not including Alfredo Morelos. The star of this season’s signing class however was Aias Aosman, who I signed from Dynamo Dresden. I believed that when he was given a chance to play a full game – which was proved by his stats with Dresden – he would perform well, and wow has he. He’s averaged 7.44, and his value has jumped up to £850k, and I now very much consider him to be competition for Fernandez when he recovers from his injury. Aosman controlled the game so well, completing 87% of his passes and scoring 3 and assisting 3 in 9 games. Looking at these fees and the return I’ve got for those fees, it’s clear to see that so far, Moneyball is certainly working. We’ve found the undervalued players, and we’ve been able to show their true value.

Screen Shot 2019-02-09 at 18.27.42

It doesn’t stop there though. The team needs to be constantly improved, and the weak spots identified. I’ve already hinted at some weaknesses I spotted above, and in the previous update. In the last update, I sold Allan McGregor to Spurs, meaning Wes Foderingham reclaimed his place as our first choice keeeper. Now, Wes went on to average 7.00 for the season, a more than respectable rating, but with Jak Alnwick as his backup – who only averaged 6.93 on loan at Scunthorpe and 6.90 when he returned – it was easy to identify goalkeeper as a weak spot. I focused on statistics as always when looking for a keeper, searching for anyone who was averaging over 7.00, and keeping clean sheets and holding onto shots. Up popped José Sá of Porto, on loan at Olympiakos, and averaging 7.05. What’s more, he also fits the Sweeper Keeper style I want from our goalkeeper more than Foderingham does. After a deal of what will eventually amount to £325k, he was ours, and will join in a matter of weeks. I’m expecting him to be fantastic next season. He fulfils rules 4, 6 and 10.

The next weakness I’d spotted was still defensive midfield. As I’ve shown above, Óscar Gil had done more than well in his position, but I’m planning to move Ross McCrorie to centre back full time next season, and I can’t always rely on Jordan Rossiter to be fit every game – although he was fit much more often in the second half of the season and proved his true value with an average rating of 7.15 – so I went into the market looking for a defensive midfielder. I say defensive midfield, but I have moved to a flat 3 in midfield with a CM-D after reading an excellent article by Keysi Rensie on this subject. This search was fairly basic, as I looked for someone making around 4 tackles and interceptions per game and averaging 7.00. I then noticed Jon Gaztañaga, playing in the Cypriot league and available on a free. When he was scouted – wisdom of crowds – my scouts absolutely loved him. With a contract negotiated for £13k a week in wages – remember, wage expenditure is much more important than transfer expenditure – he’s ours for next season and will compete with Gil for the CM-D role. He fulfils rules 2, 4, 6 and 10.

Next up was the big one. I felt that striker was a position of weakness, with Alfredo Morelos being fantastic – 33 goals in all competitions – this season, and Shankland performing well since he came in, but in reality that’s it for us upfront, so I wanted someone to shore up that possible weakness if both were injured, and possibly someone who could be a replacement for Morelos, as I’m just waiting for clubs to come in for him at this point. Now remember rule 5, strikers cost more than they should. As a result, I scoured the Bosman market and found someone who initially looked statistically inappropriate, but further investigation showed he fit the bill perfectly. That player – who remember was available for nothing – was Johannes Eggestein. He’s only made one appearance in the Bundesliga – after I signed him – for an average rating of 7.10, but it was much lower when I signed him because he just hadn’t played. However, what alerted me was that he had an average rating of 7.21 in the youth matches Bremen had obviously been sticking him in. I know for a fact Eggestein will be a huge success for us, and after using the wisdom of crowds, signed him for next season. The fact he’d made it to the end of the season still available was ridiculous, but that shows the impact of Moneyball. For some reason, the market undervalued Eggestein, and we’re going to exploit that to its fullest. Eggestein fulfils rules 1, 4, 5, 6, 9 and 10.

Screen Shot 2019-02-20 at 14.23.58

Glenn Middleton smashes in our opener against Hibs, as we go on to win 4-0 and clinch the title.

To finish this update, I’d like to do a brief case study that explains why playing Football Manager in this way is so different to how I or anyone else would normally play FM. Remember, Moneyball is all about value, and specifically, finding and using players that are undervalued in the market. However, it’s also about evaluating the value of your players that already exist at the club, and making decisions based on statistical judgements of their value. One of the key tenants to this is Rule 1:

Players are undervalued for a range of biased reasons such as age, personality, appearance. None of these matter. All that matters is production on the field, which can be analysed through statistics.

To look at this, I want to evaluate my decision based on the player Gareth McAuley. McAuley is a centre back that is already at Rangers as the game begins, and he has one season left on his contract, and he’s 38 years old when the game begins.

Screen Shot 2019-02-20 at 13.39.52

Now, logically with a player being 39, I’d ship them out and find a younger replacement if I was playing FM how I usually play it. However, with the conditions of Moneyball, we have to delve far deeper into the performance of McAuley before I make my decision. For certain, McAuley is declining in ability, but realistically, it’s only his physical attributes that are declining. His technicals remain high, and his mentals are at a very good standard for the current level of football that we’re playing – although our standard will step up next season and I’ll have to re-evaluate the majority of the team. Last season, McAuley was very much our 4th centre back with our decision to play a Back 4, and therefore only made 11 appearances all season, 6 of them being starts, as I played him in order to evaluate his performance to see what value he had left. A closer look at his season shows that he still has plenty of value, in that he averaged over 7.00 – actually 7.07 – for the season, in a good amount of appearances for his position within our squad. He also completed 100% of his tackles, and 82% of his passes. Statistically, McAuley shows he more than has value, and ignoring the effect that age has on perception of ability, I’m inclined to look to re-sign McAuley for one more reason – but only one more as his decline will come soon and I’ll have to constantly assess him. What’s more, when I open negotiations with him, he’s wanting a far lower wage than he’s on now, and when that is combined with his performance last season, this analysis shows that McAuley has more than enough value to be retained for one more season. For much the same reason that I’m going to re-sign McAuley, I resigned 31 year old left back Lee Wallace, after he averaged 7.40 across the season. On another note, I also re-signed Ryan Jack until 2023, giving him £19k per week, remembering that wage expenditure is more important and Jack averaged 7.14 across the season, and therefore more than deserved a new contract based on his value.

Screen Shot 2019-02-19 at 09.40.27

Next season, the board has been kind enough to award me a far higher transfer budget, but I’ve immediately adjusted it and moved most of it into wages. Remember Rule 3, wage spend equates to success far more than transfer expense. One of the conditions that Alex Stewart used in his own Moneyball save was not to splash out on transfers when being promoted. I didn’t include that as we’re already in the top division, but with this step up in European competition, I’m going to treat this as a promotion, and make sure that I stick with the rules I’ve given myself, and only improve on the weaknesses that I identify throughout the season.

So, that brings us to the end of the third update of the Moneyball in FM19 save. Until next time, thank you for reading and supporting this series, and as always, should you have any questions regarding this save, the concept of Moneyball, or FM19 in general, please feel free to ask in the comments section, or contact me via Twitter (@JLAspey). Thank you again, and I hope you’ve enjoyed reading the third update of this save. I’ll be back as soon as I can with the next edition, to see how we’re set for the beginning of Season 2 of this save. Thank you again for reading.

4 thoughts on “Moneyball in Football Manager 2019: III

  1. DO you pay much attention to star ability then? Or solely on stats and average rating? Interested in doing something similar

    1. I will look at star ability and how good my scouts actually reckon they are. It depends whether I’ve found them or my scouts have. If my scouts rate him, I’ll compare with stats. If I’ve found him, I’ll then ask my scouts to tell me whether they reckon he’s at our level. It’s a marriage.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s