Speaker 1: Good day everyone, Max here from Evidence-Based Football Manager. This is part 10 of my FM23 training analysis video series. There's a lot of content I want to cover today, so let's dive straight into it. Alright, this is the training schedule that you've all been waiting for. I present to you, Evidence-Based Training Schedule version 23. It's pretty similar to the alpha version of my Evidence-Based Training Schedule that I showed you in the part 8 video, but I just made some minor rearrangements to it. So let's take a look at it together. Monday – Defending, Attacking and Physical Tuesday – Match Practice and Attacking Wednesday – I left it empty because sometimes matches are held on Wednesdays Thursday – Defending and Physical Friday – Match Practice and Defending Saturday – Another match day, so I left it empty Sunday – Ground Defence and Physical So this is a schedule that's designed to cater for a 2-match week, but you can use it for a 1-match week or no-match weeks as well. All these rest sessions are deliberately placed where they are. Whenever you have an away match, Travelling uses up 2 slots in the week. The middle slot on the day before the match and also the first slot on the day after the match. So I've deliberately left those 2 slots empty in my schedule. Another thing to note is when you copy and paste a training schedule in FM23, sometimes there's a glitch in the game that kind of messes up the arrangements of these training schedules. So if that happens to you in the game, you will need to manually review your schedule and do some rearrangements yourselves. I've also prepared a separate version of evidence-based training schedule that can be used for semi-professional clubs. Semi-pro clubs in FM23 only have 4 training slots per week and I've placed the following training sessions for those 4 slots. One defending, one match practice, one physical and one ground defence. I will call this schedule EBTS23 Mini. Both the regular EBTS23 and EBTS23 Mini will be uploaded on my Dropbox folder so you will be able to download those files there. I will now show you how much player growth can be achieved in FM23 when you use my training schedule. The aim of any custom training schedule in Football Manager is to outperform the amount of player growth that can be achieved from the default training schedules that are provided by the base game. So I will show you now that it can be achieved using EBTS23. As always, I'm using my custom Test League setup for all the tests shown in this video. The Test League save file will also be available for downloading in my public Dropbox folder. The first thing I needed to do was to establish the control, so what I did was I tested and recorded the amount of player growth when the players in the Test League used the default training schedules throughout the season. Actually I have established this control before, all the way back in part 6 video I think it was, but that test was done a few patches ago, a few months ago, so I just did these tests again on the current patch, just to make sure that the control is up to date. I established two sets of controls, the first one using a save file where player injuries were deleted from the database, and the second one without deleting the injuries. After establishing the controls that will act as a baseline, it's time to test the effects of EBTS23. I applied EBTS23 to the teams in the Test League and I recorded the amount of player growth when EBTS23 schedule was used throughout the season. Again I ran two sets of tests, the first one where player injuries were turned off and the second one where injuries were turned on. I usually do 10 independent trials for all of my Football Manager experiments, which usually produces pretty reliable results, but for the tests shown in this video today, I did 20 trials because I really wanted to boost the sample size and make sure that the statistics are as reliable as possible. And let's take a look at the results. I have 4 tables here for comparison purposes. The first table is the control when injuries were turned off. The second table here is the control when injuries were turned on. The third table here is when the players used my EBTS23 schedule throughout the season with injuries turned off. And the last table here is when the players used my EBTS23 schedule with injuries turned on. Okay, it's a lot of numbers here which is a little bit confusing, so in order to make the comparison a bit easier, what I've done is to pick out a few metrics that I think are important in assessing the strength of a custom training schedule. These metrics are the amount of current ability rating that grew throughout the season, the amount of attributes that grew throughout the season, and also the number of attributes that grew by a significant amount throughout the season. So as you can see from this table, EBTS23 showed superior performance compared to the control in all of these metrics except for the growth of goalkeepers when injuries were turned off. I'm not too sure what happened here guys, it's possible that my evidence based training schedule is not as effective in boosting the growth of goalkeepers as it is for boosting the growth of outfield players. It doesn't mean that goalkeepers didn't grow at all with EBTS, it just means that the growth of goalkeepers with EBTS wasn't better than the baseline growth. Also, when injuries were turned on, the goalkeepers did show superior growth with EBTS, so there's also the possibility that the result that you see here is just due to random statistical error. For outfield players though, I think the result is indisputable. The amount of growth from EBTS23 is superior to the control in every single one of these metrics. EBTS23 is better at promoting player growth compared to the default training schedules, at least under my test league setup. Alright I will now show you some field tests that I did using my EBTS, because I need to prove that my schedule is effective in actual FM23 saves using real clubs and not just under lab conditions. For the field test, I'm using Lille and Lyon, two clubs from the French league 1. At the start of the test, for both Lille and Lyon, I used the editor to get rid of players over the age of 27, because I know from my past experiments that in FM, player growth slows down after the age of 27, so I didn't want any of those older players included in my sample pool. I first ran 20 simulations when the default training schedules were used by both Lille and Lyon. This establishes the control. After that, I applied my EBTS to both Lille and Lyon and I did 20 more trials. I recorded the amount of player growth and the amount of injuries sustained by the players throughout the season. And let's take a look at the results. I have 6 tables here, 3 tables on the left for Lille and 3 tables on the right hand side for Lyon. The 2 tables at the top are the controls, the 2 tables in the middle are when evidence based training schedule was used, and the 2 tables at the bottom calculate the difference between the top 2 tables and the middle 2 tables. For example, let's have a look at Lenny Yoro here, who is a player for Lille. In the control, Lenny Yoro's current ability rating grew on average by 8.1 points throughout the season. With EBTS23, Yoro's CA rating grew by 12.05 points. In the 3rd table, you can see the difference between those 2 values, so 12.05 minus 8.1 is 3.95, which is recorded in this table. Since my goal in these tests is to prove that my evidence based training schedule can outperform the default training schedule, what I want is to see as many positive values in these 2 tables as possible, and as little negative values as possible. And you can see that for both of these tables, for Lille and Lyon, there are more cells that contain positive values compared to cells that contain negative values, which means that my evidence based training schedule was better at boosting player growth compared to the default training schedules for both Lille and Lyon. Not all of these players grew by the same amount, and not all attributes grew by the same amount, which is expected given that this is a field test. These are all real players training in a real club, and growth in Football Manager depends on lots of factors other than training, like match exposure, professionalism, injuries and so on, so of course not everyone is going to grow in a uniform pattern. What is important is that these players on the whole grew more with evidence based training schedules than they did with the default training schedules. And for this reason, I am happy to make the conclusion that my EBTS23 was effective in a field test using Lille and Lyon. I've also conducted some field tests for EBTS Mini. I used Chester and Bath, which are two semi-professional teams from the Vanarama League in England. The test method is the same as what I did with Lille and Lyon. I first established the control by recording the growth of players when default training schedules were used. After that, I applied EBTS Mini to both Chester and Bath, and I recorded how much the players grew throughout the season. And here are the results. Again, I have two tables, one for Chester and one for Bath, and these two tables show you the difference in the amount of growth between the control and when EBTS Mini schedule was used. Since these two tables show you the difference in growth, positive values mean that EBTS Mini outperformed the control, and negative values mean the opposite. And I think I can safely say that there are more positive values in these two tables compared to the negative values, which means that players grew more when EBTS Mini schedule was used. Again, the exact pattern of growth is different for individual players for Chester and Bath, which is what you expect from field tests like these. So again, I think I can safely conclude that EBTS 23 Mini is effective for boosting player growth for semi-professional clubs in FM23. In all of these tests, I recorded the amount of player growth, and I've also recorded the amount of injuries sustained by players during the season, which is the next thing that I wanted to discuss. In the previous part 9 video, I showed you how my EBTS resulted in a slightly higher injury risk compared to the default training schedules, but that test was done using the alpha version of the training schedule. The final version of EBTS is slightly different from the alpha version, so I did the tests again in order to investigate the injury risk that's generated by my evidence-based training schedule. So let's take a look. First of all, this part here is when I used my test league save file to test the rate of injuries. When the default training schedules were used, the players in the test league got injured on average 14.8 days per season, and when EBTS 23 was used, the same players got injured on average 16.4 days per season. So the amount of injuries per player did go up by around 1.6 days per season with EBTS 23. Next we have the results from the field test using Lil and Leon. When Lil used the default training schedules, their players got injured on average 19 days per season. When they used EBTS 23, they got injured on average 20.5 days per season. With Leon, with the default training schedules, the players got injured on average 23.2 days per season, and with EBTS 23, they got injured on average 24.2 days per season. Lastly, I have the results from the field test using Chester and Bath. When the Chester players used the default training schedules, they got injured on average 6.8 days per season, and when EBTS Mini schedule was used, they got injured on average 7.2 days per season. When Bath players used the default training schedules, they got injured on average 11.7 days per season, whereas with EBTS Mini, they got injured on average 9.2 days per season. So all of these results, except for the result from Bath, show you that the rate of injuries went up when my evidence based training schedule was used. The amount of injuries went up by around 1 or 2 days per player per season, which I personally think is not a massive amount that you should be concerned about. I think 2 extra days of injuries per player is something that can be managed using various measures in the game, like squad rotations, hiring better quality medical staff, or maybe even lowering the training intensity of your team. Alright time to make a summary of what I've talked about today. 1. EBTS 23 is better at boosting player growth compared to the default training schedules that are provided by the base game. The evidence for this was acquired from both my test league lab environment and also from a field test that I did using Lille and Lyon in the French League. 2. The Mini version of EBTS 23 can be used for semi-professional clubs that have limited number of training slots per week. EBTS 23 Mini is better at boosting player growth compared to the default schedules, as evidenced by the field test that I did using Chester and Bath from the Vanarama League. 3. Both EBTS 23 and EBTS 23 Mini will raise the injury risk of your players by around 1 or 2 days per player per season, which I think is not a massive amount, but do be mindful of it regardless. Now I did want to make some additional remarks regarding my training schedule. 1. This training schedule is aimed at one thing and one thing only, which is to boost the growth of your players as much as possible. It doesn't consider other aspects of training like raising team cohesion or boosting hidden attributes. This is because the primary aim of this training analysis series has been to uncover the game mechanics related to training and using a process of reverse engineering to construct a custom training schedule. I had to isolate other aspects of the game like team cohesion as much as possible so that I could focus exclusively on training and its effect on player growth. 2. EBTS 23 is designed to be one size fits all. You can use it during pre-season, mid-season, two match weeks, no match weeks, it doesn't matter. There is no evidence to suggest that the pattern of player growth is any different just because it's pre-season. There is this popular myth among football major players that you should be doing more physical training during pre-season, which is a myth that I have a problem with, and I'm not going to go into too much detail here, but you can use my EBTS 23 during pre-season without making any changes. And also, you don't have to feel compelled to fill up every single empty training slots in your calendar. I have done some experiments in the past which I haven't published online, but I've found that more number of training sessions don't always lead to additional growth. The relationship between the number of training sessions per week and the amount of growth is not perfectly linear. So my EBTS 23 does contain some empty slots, but don't worry about them, you can leave them as they are. 3. I'm not saying that my training schedule is the best possible schedule that can boost player growth. There are 51 different types of training sessions in FM23 and 21 training slots per week, so the number of possible combinations is astronomical. And of course there could be other combinations out there that can boost player growth more than my evidence-based training schedule. But the training schedule that I offer is the best that I could come up with based on the evidence that I've gathered. Guys both EBTS 23 and EBTS 23 Mini will be uploaded in my Dropbox folder, link provided below. Download it, try it and let me know how you go with them. Leave some feedbacks in the comment section below and I'll reply to them as much as I can. Everyone thank you so much for all the support, you guys are the best. This training series is not over by the way, there are a few more game mechanics related to training that I wish to investigate a bit further, so this training analysis series will continue, but yeah at least the training schedule is done. Okay guys, thanks again and I will see you all next time.