This Week in the PNFL Simming

[Jerry-Redskins]

Great discussion guys. It was a lot of fun this week. I found it interesting as I made the decision after internal discussions with myself about 8 seasons ago. I decided to go no sim. I now use a pool of the generally most successful plays. I study my opponent and anticipate what they will do. I then use the plays that counter that best. I believe I could alter my PPP a little more each week, but I do change 25% or so of the plays each week on average. Nothing like the Cardinals, but that amount of change and the fact I use larger numbers in each play category gives me enough variety. I think simming absolutely narrows the unknown if you game plan the opponent correctly. If you have the best roster and good situational play calling, simming will improve your odds a lot. Looking at you Rich

I have proved however that analysis of the opponent and guessing what I will see can be pretty successful. Most games are decided by only a few plays each week in my opinion. The better player in the situation and the right situational play call will win more of those key plays. My goal is to put myslef into those situations the best and force my opponent to worry about me more.

[Rich-League Officer]

If that works for you, keep doing it.
Simming works but there is more to it than just simming and parsing the data.
It's something I have told many coaches over the years, that more than simming, the most important skill is scouting your opponent.
And I do think it's a skill going beyond just looking at the log for what that team did last week or last few weeks.
I would say the greatest skill I had was predicting what my opponent would do.
It comes from recent history but it also comes from past history in a season and past meetings.
History of what coaches did vs higher level opponents vs lower level opponents.
It comes from years of experience coaching and watching what other coaches like to do.
It comes from testing the plays so often that you know which plays stink and which don't.
You need to wrap your head around all of that to predict what a great coach will do.
Only then, does the the simming matter. And even then, the great coach against you may win the game.
And scouting your opponent goes beyond the plays. Recreating his profile in as much detail as possible.
What specials he uses, how often he's going on 4th-1, how often does he trick you and run on 3rd-9?
And once you have that, you need to make your profiles are as bulletproof as possible.
Are my timeouts accurate? am I foolishly using them only in the last minute of games? maybe only on offense?
Will my offense flow? and what does that even mean?
I know my runs are getting 4yds per but what if my short middles are routinely getting 5yds per. Am I finding myself in lots of 3rd-1?
Do you ever watch your game and think, shit, another 3rd and less than a yard or 4th-1?
Stalled drives because the plays you selected are not allowing your offense to flow.
And maybe on defense, you love sacks, but the flip side is big chunks against.
So there are games where its stop, stop, +35 and you are giving up huge chunks all day yet his QB is barely 50%
You as OC/DC need to coach that out of there as much as possible.
Trust me, coaching this game is a lot more than just simming plays.

As Mitch told you, he simmed but his data was garbage because his scout plan for Pittsburgh was wrong.
Garbage in, garbage out. Simming can hurt you too.

[Mitch-Oilers]

ditto everything Rich said...

Other things you have to consider as well are sacks and turnovers when picking offensive plays.

You may see a PML play that averages 55% comp% for 15+ yards. However, if the sack% is 25%, that means 25% of 2nd downs will be 2nd & 10+. Talk about messing up your flow.

[Rich-League Officer]

There are dozens of other thing to consider.
How you use your 2 point conversions for one.
James made some arguments for it but he's going strictly on analytics.
Football and baseball (most sports) are situational. Going for 2 will work in the long term if you sim 10,000 games.
But does it work on this play right here and now in the Super Bowl?
Using them smartly will make you a better coach. Using them 100% of the time will bite you in the ass sometimes.

When you go on 4th down is the same thing and that works hand in hand with your ability to execute and pick plays.
I am seeing a team or 2 that already struggles making first downs suddenly go on 4th more often.
This is bad coaching. This same coach has profile errors for not going for 4th down when he should.
Also not calling time outs properly or at all.
Coaches routinely are making mistakes in their profiles which definitely cost them chances at winning.
The best coaches do not make these mistakes.
As I said, it's more than just simming games.
You could coach, never sim a game, use the logs, logger and other tools, fireproof your profile and win games.
It's possible.

[Matt-Jacksonville]

Thanks for the insight Rich. I think this is what I'm missing. Working with data all day in my day job for the last several years, I think I've leaned too much on data and analytics and not enough on actual "coaching" as I'd like to put it. Also, I'd like to add, some of this takes time in the game to develop. Watching film, figuring out how stuff works, and making sure you aren't hurting yourself. Some of us have been doing this long enough that we have a process down where it doesn't take much time. Others may still be figuring some of this out or may have forgotten some of it.

[Matt-Jacksonville]

I'd also like to add that it also depends on what you are looking at in your sims. Are you looking at ball distribution and play calling(testing profile) or are you looking at what individual plays do ? I would say simming can be a good thing if you are testing your profile and trying to iron out bugs.