Position Training Now Impacts Tactical Familiarity

[Joey Numbaz]

I had not seen this discussed before, if I am duplicating it I apologize.

Looks like positional training has an impact on tactical familiarity. This seems very important for those of us that like to train positional versatility. I will often train a player at a position he isn't necessarily playing at the moment, so he can provide cover, or move there in the future, etc.

I am still not sure of how long it takes, but I noticed this:

Michael Keane has been on my team since the beginning of the game in July 2017. It's now October 2020.

Matthijs de Ligt was just signed this summer. But he's more tactically familiar with my system than Keane? Say what?

Specifically look at this for Keane:

And de Ligt:

Take note of the "Position/Role/Duty" bar for each player.

I have been training Keane as a DM, just in case I need to move him there at some point for some reason. I like to have options. Check this out. Despite being Awkward as a DM, he is nearly perfectly fluid (9500/10000) as a DLPd for "Position/Role/Duty". He's 9000/10000 as a DLPs, and my tactic does train D more often than S, I tweak it sometimes.

Also note, my tactic doesn't have a BWM in that position (or in any position), but he does have the same tactical familiarity with that role too - as a DM (9500 defend, 9000 support). As a MC, which he is Competent at, his familiarity is roughly the same as he is a DCx.

I actually think this is a great addition, or maybe it isn't an addition and has been around awhile and I never noticed it.

The next step is to try to figure out how long it takes once I get his training back to BPD. From what I can tell the role you train the player as doesn't matter in terms of tactical familiarity, just the position.

So yeah, the duties you have in your tactic, and the positions you train your player at have an impact on his positional familiarity. It should probably be called Position/Duty familiarity, as I don't see any impact of the role. EDIT - see the third post in the thread, there is role impact. Keane has been training as a HB, which I don't use, just because it's the combination of the broadest set of attributes and most similar to those of a BPD, and his main position is DC.

[Joey Numbaz]

Adding a note here for research purposes, changing Keane back to a BPD in training on 10/23/2020.

His current familiarity:

DCRx 5032/10000

DCLd 4528/10000

Those are the two roles in my tactic. He almost always plays on the right as the stopper.

DCRd 4931/10000

DCLx 4528/10000

So I guess there is some impact based on the role they play in the games (or that you practice in your tactic?) and even the side of the field they usually play on. This is pretty excellent from a realism perspective if you ask me.

But obviously the position you have them training at is the main driver of this. There's no way he'd be 9000-9500 in the DM position and 4500-5000 in the CD position if that wasn't the case.

 

[Joey Numbaz]

This is strange though ... I've been training de Ligt and Mbemba as CWB. And their tactical familiarity does not seem to have suffered.

Perhaps Keane is taking the hit as a CD because he's awkward as a DM so he's putting all of his energies into learning the position?

Mbemba has no familiarity at all with WBR, he's not even awkward there, but de Ligt is still unconvincing. So I doubt that's it.

Going back to September 30 (I save every few days).

Keane:

Pos.  -  9/30     10/23   net

DCRx - 5842    5032   -810

DCRd - 5725    4931   -794

DCLx - 5257     4528  -729

DCLd - 5257    4528   -729

He's losing ground - 23 days later, despite playing this position exclusively, he's worse.

DLPd - 9500    9500    0 (is he maxed out? maxed out for 'awkward'?)

DLPs - 9000    9000    0 (ditto)  

Mbemba (who plays Sweeper and trains as a WBR/CWB):

Pos.  -  9/30     10/23  net

SWd - 8800    8620    -180

Ls -     8625    8719    +94

La -     8360     8897    +537

Note, somewhere in that month, I did change the role in the practiced tactic from SWd to La. Although in game I will tweak it based on the game situation.

WBRa - 7812   8844     +1032

WBRs - 7972   9025     +1053

We practice and play the position almost exclusively as a WBR/CWBa, although I will sometimes drop it to support very late when closing out a win. Mbemba did move over there late one game, and very well may have played as support.

De Ligt (plays DCLd, trains as WBR/CWB):

Pos.  -  9/30     10/23  net

DCLd - 8897     8842    -55

DCLx - 8897     8842    -55

DCRx - 9886    9825    -61

DCRd - 9832    9750    -82

WBRa - 8780   9659  +879

WBRs - 8960   9857   +897

Another wrinkle! Both De Ligt and Keane it turns out "prefer to be on the right in a two-man pairing". So this is where that matters. The get a 9.0-10.5% bonus for the preferred side in these examples.

[RocheBag]

It has nothing to do with them training positions. Their familiarity with the role you put them in affects the tactical familiarity.

[Joey Numbaz]

46 minutes ago, RocheBag said:

It has nothing to do with them training positions. Their familiarity with the role you put them in affects the tactical familiarity.

Except it doesn't. Or more specifically, it does a very little, and that little bit is greatly outweighed by the position the player has recently been training at.

How else to you explain Keane having much more tactical familiarity at a position he has never played for me, but has been training at for awhile, than the one he's been playing for 3.5 seasons?

[craigd84]

where are you seeing those numbrs?

 

[Joey Numbaz]

So I just played a game (Saturday). Before the game, I switched Keane's training from HB to BPD, after the game his familiarity as CDx went from 5032 to 5200.

His familiarity at DLPs went from 9000 to 8962.

Day off the next day (Sunday). At 9 a.m. his DLPs went from 8962 to 8924. DCx dropped from 5200 to 5158.

6 a.m. the next day (Monday), DCx down to 5116; DLPs 8886.

Note, I have match training at 10% - the first click over from the left most one. For this match I am training attacking set pieces.

6 a.m. the following day (Tuesday), DCx is down to 5074; DLPs 8848. There is a game today, and half of the day (per the training screen) shows as match prep. Keane is going to sit this game out, it's a League Cup 4th round game against one of the Championship's relegation candidates.

8 a.m. Wednesday after not playing in the game, Keane is down to 5032. Back to where he was before playing Saturday's game, exactly. DLPs is 8811.

Interesting. So now I'm starting to think the training component of positional familiarity is a combination of:

1) the position you are training

2) how much of your team training is match training

3) how much of your match preparation is "Match tactics"

There's a Sunderland game coming up Saturday (yeah they are back in the EPL in 2020-21, but in 20th place). I'll max out match training starting Thursday (off day Wednesday). Choose Match tactics as the main focus as well.

Thursday, 6 a.m. after the off day, Keane is at 4990 CDx, 8773 DLPs.

Friday, 10 a.m. after a day with the bar all the way over to Match Training, with match tactics as the focus, he drops again to 4948/8735.

Saturday 8 a.m., 4906/8697. Strange to me that he's losing at a faster pace (42 points per day) as a DC than as a DM (38 points per day).

I am about to play the game ... anyone want to guess based on all of this what his familiarity will be after the game?

My gut was that he will gain more than the 168 he gained in the last game. I was basing this on the fact that he is practicing that position and the team practiced match training at max and match tactics as the focus as well. But the benefit isn't seen until after the game.

The only other wrinkle is that since I didn't change the match training number % until Wednesday, while it shows three days of match practice on the team training calendar at first, it didn't take. Looking at it again it shows less matching prep time this week after all.

So now I'm guessing he'll gain roughly 168 points from this game, and then next week, his points will start to pile up during training instead of dropping. The problem is that I have a game with Leverkusen in the Champions League and I'll need the boost from one attacking movement or attacking set pieces for that one. I guess I can run the sims with match tactics to test for the days leading up to the game, and then go back and run them again to actually play it.

[Joey Numbaz]

30 minutes ago, craigd84 said:

where are you seeing those numbrs?

 

Player page -> Development -> Tactics

It's on the far right.

[Joey Numbaz]

12 minutes ago, Joey Numbaz said:

The only other wrinkle is that since I didn't change the match training number % until Wednesday, while it shows three days of match practice on the team training calendar at first, it didn't take. Looking at it again it shows less matching prep time this week after all.

Yeah, definitely didn't take. During the game, my assistant is talking about how our focus on attacking set pieces is really paying off.

As a side note we've scored 3 goals off set pieces in the first half of this game, another countering their set piece and hit the cross bar on another corner. 

[Joey Numbaz]

Keane was the Man of the Match with a 9.0 rating. We won 5-1, I would have subbed him or De Ligt at halftime, but science.

Immediately after the game, Keane's familiarity has run up to 5116 a gain of 210 points. So more than he gained after the game a week ago. He is still behind the 5200 he was at after last Saturday's game though. So to summarize, he gained 168 points in the last game, lost 294 points during the week (42 points per day, whether the team was off or not).

So with match training very low, playing 90 minutes once a week is not enough to build familiarity it appears.

De Ligt, who I didn't check daily, played the same as Keane. Skipped the midweek game, played Saturday. He is at 8912 as a CDLd. He was at 8842 before the game a week ago. So he has against 70 points since then, playing 2x. Keane has gained 84 in an apples to apples comparison.

Sunday morning, Keane has lost another 42 points. Down to 5074. De Ligt on the other hand, only lost 32 points, he's at 8880.

Monday morning, Keane is down 5032. Another 42 points gone. De Ligt down to 8842. He lost 38 points today.

But how about WBR/CWBa for De Ligt? Remember way back when (10/23) he was at 9659? He has not played a second there. He's at 9800. How else would he be gaining familiarity if training position did not impact that?

Meanwhile Keane, who stopped training as a DM, has dropped from 9000 to 8622 during the same time period.

Tuesday a.m. Keane is at 5111. Ah hah!! That match training slider (and perhaps match tactics as the focus) does matter for this. For the first time, he gained on an off day. He improved 79 points, from one day of team practice at maximum Match Training/Match Tactics focus.

De Ligt however? He's at 8838. He lost 4 points at the position he plays. At WBR/CWBa ... there he's still at 9800.

OK, resim, of Monday. This time I choose Attacking movement as the focus. Still maximum match training. Now Keane turns up at 5128 familiarity. He gained even more points (96 vs. 79) than with Match Tactics. That's strange. De Ligt remained at 8838 and 9800.

OK, resimmed again with the team main focus as attacking set pieces, and Keane ends up at 5089. Last time, I kept the main focus as Match Tactics, but changed the match specifically (on the calendar date itself) to attacking set pieces. Interesting. Let's try this again. Is it random within a range? Or is does it matter where you change the focus? De Ligt is at 8839 and 9800 now.

Resimmed it as main focus match tactics, but specific game as attacking set pieces, Keane comes out at 5094. De Ligt 8832. I am starting to think there is a random element for how well practice went, within a certain range.

Resmimmed again with those exact settings and Keane gets 5136. De Ligt 8851. So it definitely appears to be random with a range, and it looks like the whole team moves in a certain direction within that range. And that range is influenced based on your match training sliders and focus it looks like.

Another note, the best I can tell at this point, is maybe there is a lag from when you start training at a new position before you start gaining the familiarity points from training days. EDIT - I don't think this is true. Based on further testing described below, I think changing the match training slider doesn't take effect until the following Monday.

Because De Ligt kept gaining at WBR and Keane did not at CD even before tweaking the team training.

OK, another thing to test, remember I've been training Keane as a ball playing defender, despite the fact that he plays central defender. Maybe that has an impact?

Going back to 10/23 before the game Keane was:

CDR/BPDx - 5032

CDR/CDx - 5032

Before game on 11/3

CDR/BPDx - 5136

CDR/CDx - 5136

I think that shows the position you practice matters, but the role does not.

[Joey Numbaz]

Played another game Tuesday, Keane picked up 210 more points (up to 5346). So it's only a couple of data points, but he added 210 points both games where he started the day training as a CD and playing it in the game. He only picked up 168 points the day he started as a DM (I changed him to CD training before the game, but after the day started). For whatever that is worth.

De Ligt improved from 8851 to 8993 (+142 points). De Ligt remains at 9800 as a WBR/CWBa. Meanwhile Keane is down 8584 as a DLPs.

[Joey Numbaz]

Ok, Wednesday is a team off day, and Keane lost another 42 points, down to 5304. De Ligt is down to 8955, a loss of 38 points. I think we've pretty consistently established what happens when they have an off day, or have very low team match training at this point.

Thursday, with normal practice scheduled, including full Match Training slider and Attacking Set Pieces, Keane gains 17 points to 5321. De Ligt lost another 37, down to 8918. This has to be because he's training at another position. I cannot think of any other explanation. His WBR/CWBa seems to be maxed out at 9800. WBR/CWBs he's a perfect 10000. He was 9857 back on 10/23. It's now 11/5.

There's an international break coming up. Very curious to see how that impacts things. Keane and De Ligt are both internationals who will leave for a week and a half.

[Joey Numbaz]

It's now November 18 and Keane is back from the international break. He's currently at 6640 for a CDR/CDx. So he did gain ground while away. But I assume Gareth Southgate also used him as a CD while he was gone. De Ligt is now 10000 at CDL/CDd.

As a WBR/CWBa De Ligt is now 9049, down from 9800. But right before he left, I switched his training back to BPD. He's a 9234 WBR/CWBs.

 

[Joey Numbaz]

And the experiment is over. Keane is out for 5-7 weeks with a knee injury.

Actually I guess this will let us take a look at what an injury does.

He leaves for his injury at 6640 for DCR/CDx. DM/DLPs he's at 7780.

Bummer. He signed a contract extension yesterday. But I digress.

 

[Joey Numbaz]

Interesting. I guess that while Keane is injured he is heading into the film room every day. As of November 24, after 6 days injured, Keane's tactical familiarity as a DCR/CDx is up from 6640 at the time of injury to 6930. That's pretty sweet. DM/DLPs is down to 7552.

[RocheBag]

On 30/07/2018 at 14:18, Joey Numbaz said:

How else to you explain Keane having much more tactical familiarity at a position he has never played for me, but has been training at for awhile, than the one he's been playing for 3.5 seasons?

His attributes are more suited to that role than the one you've been playing him in.

[Joey Numbaz]

16 minutes ago, RocheBag said:

His attributes are more suited to that role than the one you've been playing him in.

Except that once I stopped training him as a DM his familiarity there has been dropping like a brick. I've chronicled it nearly day-by-day. I don't see how you can reach any other conclusion.