CPU Bullpen Management
Collapse
Recommended Videos
Collapse
X
-
-
Re: CPU Bullpen Management
After reading this thread and playing with the sliders myself, I'm convinced that the stamina slider is really for effectiveness. The higher the stamina slider, the more effective the pitcher is, and the higher the hook slider, the less the manager cares how effective you're supposed to be and more about how you're actually performing. This might be all semantics, but this makes sense to me.
A higher hook slider gives the Mgr more confidence in his pen, and possibly less confidence in his starter. With higher hook, I noticed fewer starters going 6 IP/game. 5 was often a stretch. My theory is Hook at 0 makes it absolutely about the situation. Possibly similar to setting dynamic difficulty to 0 to lock in.
So I've done a lot of testing on this in past iterations and what I discovered is that the CPU cares most about effective pitches IE the point where a pitcher starts losing effectiveness on their primary pitch (assuming it's a fastball, the rate is different for other pitches).
I found with stamina at default or lower, the CPU would rather leave a tired guy in to maintain their bullpen's effective pitch count then make an effort to win the game. Their actions indicate their logic is: well this guy is already past the point of effective pitches and I only have x number of effective pitches left in the bullpen. Might as well run this guy ragged and save my effective pitches for tomorrow.
I found by raising pitcher staminas, both starter and reliever, and then raising the manager hook solved the problem... Mostly. There was no setting I could find that made me completely satisfied with bullpen management, but I did manage to close the gap significantly. Point is, the CPU is far more concerned with pitcher performance and situation then maintaining total bullpen pitch effectiveness on my settings.
My current settings on 20 are all pitcher staminas at 7, CPU manager hook at 6. I'm still playtesting my sliders in 20 but I'm confident you'll see an improvement in bullpen management with these settings.Comment
-
Re: CPU Bullpen Management
I did watch a game earlier with my sliders, but didn't get as much bullpen action as hoped. Kept notes on pitch count, hits, walks, etc. Might watch another and hope for better.Last edited by garry1221; 04-30-2020, 04:07 PM.Comment
-
Re: CPU Bullpen Management
After reading this thread and playing with the sliders myself, I'm convinced that the stamina slider is really for effectiveness. The higher the stamina slider, the more effective the pitcher is, and the higher the hook slider, the less the manager cares how effective you're supposed to be and more about how you're actually performing. This might be all semantics, but this makes sense to me.
Higher stamina means it will take a longer period/higher pitch count to cause a pitcher to become less effective.Comment
-
Re: CPU Bullpen Management
I'm not sure for played games as I am primarily a sim GM. I do think that lower sliders = more runs and higher sliders = more runs so it really just matters if you want the game to be more pitcher or more hitter friendlyComment
-
Re: CPU Bullpen Management
That is correct, but it does also control actual energy. You'll notice that with a lower number for starter stamina, depending on pitcher, he may be less than full energy for his next start. Raise it and you'll see the opposite.
Higher stamina means it will take a longer period/higher pitch count to cause a pitcher to become less effective.Comment
-
Re: CPU Bullpen Management
Yes, it does, but I think the way it goes about this is somewhat of a self-fulfilling prophecy. It takes longer for a pitcher to become less effective(than their baseline talent) without it because they're *more* effective(than their baseline talent) with it. The stamina slider boosts both their stamina AND the benefits of that stamina in regards to their effectiveness. Perhaps having more energy means each individual pitch takes less energy so therefore you can put more into them? I can't tell if I'm making the simplest stuff the most complicated or if I'm on a verge of a breakthrough lol
Gerrit Cole is a bonafied ace. When simming, I have starter stamina at 6 (or 7, I can't remember which). By this logic, he would be more effective than he usually is, especially considering his high stamina. However, he has an ERA of 5+ near the end of April in my franchise.
On the flip side, in a previous franchise, I had starter stamina at 1. Through the same time period, Cole was dominant as you'd expect.
There will always be variance, and those are tiny sample sizes. But you would expect the variance level to be lower in scenario 1, if this theory was founded.Comment
-
Re: CPU Bullpen Management
Based on my playtesting, the effect of increasing pitcher stamina is more like flattening out the early curve, with a sharper regression of pitch quality near the end. The end is pushed out, but only a minor amount.
Sent from my Pixel 3 using Operation Sports mobile appComment
-
Re: CPU Bullpen Management
I don't think it's that deep, and let me give you an example:
Gerrit Cole is a bonafied ace. When simming, I have starter stamina at 6 (or 7, I can't remember which). By this logic, he would be more effective than he usually is, especially considering his high stamina. However, he has an ERA of 5+ near the end of April in my franchise.
On the flip side, in a previous franchise, I had starter stamina at 1. Through the same time period, Cole was dominant as you'd expect.
There will always be variance, and those are tiny sample sizes. But you would expect the variance level to be lower in scenario 1, if this theory was founded.
At the end of the day, it's still baseball. The reason why a pitcher with all 99s still won't throw a perfect game every time out. Cole also might've been "cursed" that particular season(I do believe that certain players throughout a franchise will be doomed to have terrible seasons at random, but that's another story). I see about a 1.2 run differential in team ERA across the league if I sim with sliders at 1 vs. 6-7. It's just a little nudge; a fastball that comes in at 94 instead of 96 or a breaking ball that hangs just inside the zone because the pitcher was at 40% instead of 45%. The CPU/sim engine adds that into their RNG for every pitch in every game, which influences the entire league one way or the other. We know that these sliders affect the game, we just don't know how exactly it comes to those conclusions....
But you're probably right, it's not that deep 😅Comment
-
Re: CPU Bullpen Management
I decided to pair up teams based on league and OVR using the latest live roster (NYY vs LAD is the exception). I set their pens based on pitch clutch. I also set their lineups using the same criteria for each. I'll use SP's closest in OVR. All this is to hopefully achieve as much parity as I can before letting the cpu do its thing.
I'll be taking notes on IP, situation leading to SP being pulled, pitch count vs STA, when I feel I see a pitcher losing command, etc. When I have a few games worth of data i'll post my findings.Comment
-
Re: CPU Bullpen Management
So this is what I do, I have manager hook on 0 and go in team select to switch over to the team and put in the reliever/closer myself and don’t cheat for myself.
That is the only way IMO to fix this issue
Sent from my iPhone using Operation SportsComment
-
Re: CPU Bullpen Management
Have you tried him with the stamina slider at 9? I hear what you're saying, but it's not a fair experiment to compare a +1 or +2 boost to a -4 cut.
At the end of the day, it's still baseball. The reason why a pitcher with all 99s still won't throw a perfect game every time out. Cole also might've been "cursed" that particular season(I do believe that certain players throughout a franchise will be doomed to have terrible seasons at random, but that's another story). I see about a 1.2 run differential in team ERA across the league if I sim with sliders at 1 vs. 6-7. It's just a little nudge; a fastball that comes in at 94 instead of 96 or a breaking ball that hangs just inside the zone because the pitcher was at 40% instead of 45%. The CPU/sim engine adds that into their RNG for every pitch in every game, which influences the entire league one way or the other. We know that these sliders affect the game, we just don't know how exactly it comes to those conclusions....
But you're probably right, it's not that deep 😅
if you also have hook low, the manager dosn't care about stamina, so he'll let the starter roast longer than he would normally. Add to the fact that starters will begin starts without full energy more often, and boom, higher team ERAs across the board.Comment
-
Re: CPU Bullpen Management
Cy, took a look at your TrueSim sliders last night... In games you play, how is the CPU bullpen usage? I think you have things setup as 7/1/1.
Honestly, I'm just trying to find a good setup for games I play vs the CPU... I'm going with your Sim sliders to simming games.Milwaukee Brewers | Green Bay Packers | North Carolina Tar Heels | Wisconsin BadgersComment
-
Re: CPU Bullpen Management
Cy, took a look at your TrueSim sliders last night... In games you play, how is the CPU bullpen usage? I think you have things setup as 7/1/1.
Honestly, I'm just trying to find a good setup for games I play vs the CPU... I'm going with your Sim sliders to simming games.
I've gotten great results.
EDIT: It's not 7/1/1. I've been using SP stamina at 8, RP stamina at 2, and hook at 3.Last edited by Cycloniac; 05-01-2020, 11:03 AM.Comment
Comment