GG, Ioan? If your experiment was to riddle my team with injuries, you did accomplish that... I'd be interested to know what you're trying to figure out.
Easy thing. I found out that not all plays are common anymore to all roster constellations in v0.46 (e.g. FL Hitch, HB Flare, HB Blast etc). There might be very different plays which work with a specific roster. So I am experimenting with ALL plays in defense and offense to find out which plays suit MY roster. This will last for round about 1,5 to 2 seasons.
In one league I tried the best plays out of Big Data League, but it didn't work with my roster in that league. I ended up with a losing season.
In v0.45 you had some common plays which always worked, but v0.46 seems to work more acc. skills in your roster.
I was wondering why Wrexham got so bad all of a sudden in the UK league. Not to turn this into a coaching philosophy thread, but won't it be hard to build up play familiarity if you're trying *every* play? Shouldn't there be a little consistency (even 5-7 straight games with the same plays) so the players get better over time?
The minimum for each play is 10 times played. So I get a little familiarity and a clou how plays work. It's a tedious endeavor, but it seems to be working.
Re: 2038 good game
by
norm
@
2/05/2022 11:41 am
One warning about this strategy: play knowledge is tied to your players. So if you spend a whole year learning plays, but then decide to let a bunch of people go, or they retire... you could be back at square one next season.
One warning about this strategy: play knowledge is tied to your players. So if you spend a whole year learning plays, but then decide to let a bunch of people go, or they retire... you could be back at square one next season.
I was just looking this morning at my players, skill mostly, for their individual play knowledge. This is a good point and glad you mention this.
One warning about this strategy: play knowledge is tied to your players. So if you spend a whole year learning plays, but then decide to let a bunch of people go, or they retire... you could be back at square one next season.
I was just looking this morning at my players, skill mostly, for their individual play knowledge. This is a good point and glad you mention this.
Interesting. I would have thought you'd need more play repetitions to build up familiarity. That's why I figured teams that only used 12 offensive and 4 defensive plays did so well because their players had run them so many times they could execute at an elite level.
Re: 2038 good game
by
norm
@
2/05/2022 5:31 pm
dangalanti wrote:
Waitwut wrote:
norm wrote:
One warning about this strategy: play knowledge is tied to your players. So if you spend a whole year learning plays, but then decide to let a bunch of people go, or they retire... you could be back at square one next season.
I was just looking this morning at my players, skill mostly, for their individual play knowledge. This is a good point and glad you mention this.
Interesting. I would have thought you'd need more play repetitions to build up familiarity. That's why I figured teams that only used 12 offensive and 4 defensive plays did so well because their players had run them so many times they could execute at an elite level.
My unprofessional opinion is that running a play 10 times isn't enough to gain good knowledge. Not sure what the right number is, but I would guess it is more, based on how slowly that play knowledge bar seems to move for me. That said, I believe the new engine has greater penalties for the 12 plays only offensive playbook - that is, using one play over and over again will lead to the sim stacking the deck against you more and more. This happens even if the text doesn't say "the offense is using that play too much." But of course on a sliding scale. Again, I'm not sure what the breakeven point is. Five times? Seven times? No idea. But the intent is to reward diverse play calling and penalize running the same thing over and over - but even with 12 plays, if things are equal, you run everything 5 times a game - that might not be too bad. Not sure. You will have to ask the true number crunchers to find out.
