Taking one for the Team (The Series)

[motnahp]

I've addressed this issue before, but I don't know what has been done to fix the problem. Running 1.67 Beta in a 2006 replay assn, using actual MLB daily rosters.

The AI-managed teams are still occasionally sending out some poor soul to the mound to receive extreme abuse at the hands of the other team's batters. I'm attaching a screen shot of the box score for the latest one. I don't catch them all, though. I wonder how many times this is happening.

You'll notice that the AI did (eventually) bring in another pitcher, who completed the last two innings. My question is this: where was this pitcher in the 7th inning when his teammate was giving up all those runs? Someone has got to come to the rescue. A fifth starter, the closer, a worn-out reliever, a backup middle infielder....someone.

[motnahp]

Looking through my "bottom ten" in assn ERA, I only found one other case of this. In this one the Pirates were on the offense and received a freebie. This is the first time I've seen a starting pitcher with this anomaly. Usually, its a reliever.

[BleacherBum]

I started a fictional league in 1950, and I see pitchers getting beat up for 10+ runs all the time. I think for my league its related to the time period, and the game trying to minimize the use of relievers. I'm not sure this is realistic either. I would think that even in the 1950's, if a guy was getting pounded, he came out of the game. I need to do a little research though to go beyond just an opinion.

[KG Erwin]

What are the pitching ratings for the guys getting bombed? Obviously, if the talent pool is thin, and a guy with 5 stuff/velocity gets thrown into the lions' den in the majors, he's gonna get beat up.

Is this what's happening to your guys? How large are the rosters? 50 players or more?

OK -- I looked over my Replay 1947 association major-league pitching staffs, and while several guys have velocity ratings of 5, NO ONE has a stuff or control rating in the single digits.

So, how in the world does a scrub even make it into the majors?

[motnahp]

35-man rosters. I'm mirroring the real MLB rosters as they stood on each day. Most of these guys in question are 5-pointers. I recall that a PS release from about a year ago resulted in seldom-used batters and pitchers becoming 5-pointers (on the 1-100 scale). I didn't pay it too much attention at the time, but this reduction obviously was a bit of overkill, IMO.

As I've mentioned in other posts, a 5-point batter won't have too much of an impact on the grand scheme of things. A 5-point pitcher, however, can skew an entire association's stats. At some point in time, I'll fool around with the ratings and try to determine what the "ideal" minimum should be. If a pitcher was on a roster long enough to have more than one appearance, he obviously had SOME skill, much more than what I'm seeing with these guys walking 15 batters in an inning and giving up 15-20 runs in one appearance.

As I mentioned in another post, I'm fully in favor of the September call-ups. Guess what, though? My association is currently on August 31st and I fear an influx of these 5-point pitchers and an increase of games where teams score over 20 runs. Come to think of it, this would be the ideal place to save a copy of my current association, so I can tinker with those 5-pointers and make my comparisons. I'll post the results here some time in the future.

[jeremy7227]

I agree with you on 5-point players. This is one main reason why I have suggested that a utility -in game or third-party- be developed that would allow us the ability to make bulk ratings changes on a filtered pool of players that meet a certain criteria. I recently started a new association using real players and I find it takes hours to visit the pitching pool, apply a filter and then edit bios player by player to modify the 5-pointers. I understand why, in my league, they are created this way but for better playability I would love a utility to mod a group at once.

[KG Erwin]

2006, motnahp? There's you're first problem. You started with real players, and then have generated fictionals to follow? Nah, that's not gonna work.

The historical model, as I've found, must start within a pre-existing time frame -- say, from 2000 to 2006. In designing templates, the furthest back one can go is the early 1920s.

Now, let me tell you guys this -- if you wanna play with real players from 1920 or so forward, you'll have a wonderful time with the game.

[puresimmer]

I don't understand what you are doing to have so many 5-point players?

Can you give me the exact steps to create an association that illustrates this? The only thing I can assume is you are using real players but the pool of real players you are using is not large enough to fill whatever association config you have. Since PureSim trys to "keep it real" when using real players (e.g. not generate a bunch of fictional guys that play instead of whatever real guys you have) it is filling out the gaps with scrubs.

Am I making the right assumption?

[jeremy7227]

Actually no. The real players in my association are from pre-1900 and so pitchers have low walk and strike out numbers relative to their very high IP. This seems to translate to "5" (out of 100) velocity more often than not. This is true if I start in Sandbox mode with an empty association or in PS Classic starting with a Fictional Pool and importing real players. The scrubs aren't the issue for me at all - they are usually SO bad that you won't have to worry about them except maybe a cup-of-coffee in Sept. However the real guys from the deadball era have very good numbers in their other ratings and so they are getting knocked around unfairly.

There is also a problem for the batters from this era. It just doesn't manifest itself as dramatically. Because of the low number of walks the batters EYE ratings are single digits to mid-teens when they are generated. The game doesn't "normalize" these players to the association year. I don't think it should necessarily do that but the "5's" are a little too plentiful.

Weird things in Sandbox mode happen on import too. I imported Cap Anson from 1871 and 1872 to compare and he had a "5" contact. He was the only player in the pool. In PS Classic this doesn't happen (this is why I moved my association concept there and accept the fictional guys). In those leagues HOF hitters like Cap Anson and Jim O'rourke come into the league with mid 60's contact which I think is low but acceptable.

[jeremy7227]

SBox mode is cool I think. Or rather the more you import the more it will look like PSClassic. I imported more players and noticed that the ratings fluctuate until the pool is set. This way the players are rated relative to the overall talent of the association not an outside factor. This is cool.

Here is a screen shot of the 1871 pitcher pool imported into a fictional league starting in 2000. I don't doubt the accuracy of the rating (these guys did throw underhand afterall) but to get results near what one would like to see across an association and similar to these guys real life effectiveness there needs to be a way to lift players to a usable standard.

[jeremy7227]

Here's Al Spaulding with his real stats:

[jeremy7227]

Here's Cap Anson in the same association. Not as bad as Spaulding and he may perform ok. Like KG said- the batter isn't going to skew the whole leagues results.

[jeremy7227]

One more thing on this and I'll shut up. SandBox Mode does a really good job with batter ratings since you can import into an empty association. The problem is that SBox stopped respecting "Allow Duplicate Players = N" in version 1.66Beta.

[jeremy7227]

And the result of the preceding import...

[motnahp]

We apparently got our wires crossed somewhere along the way. My assn is strictly real players, with zero fictionals. This assn's rosters are mirroring the ACTUAL rosters each given day. The 5-pointers are actual MLB players, mostly rookies. Some were pretty bad in their actual MLB stats, but quite a few of them were very respectable. They just didn't have enough appearances/at-bats to import as other than a 5-pointer.

I know I'm skirting the edges of this program's capabilities, but if I'm doing a replay, I feel that even the seldom-used players should perform much closer to their real-life stats. Just because they only had a few appearances in real life is no reason for them to take the mound and give up 20+ runs in one appearance. As it stands now, these guys are performing worse than ANY major league should be performing. I'm not looking for the players to have EXACTLY the same results. That would be no fun. There, I've vented, and feel better now.

[motnahp]

This assn is a 2006 replay, with financials, AI trades, injuries, and affinity OFF. I actually have a list of all real MLB transactions from 2006. I use the actual transactions each day to make every team reflect what the rosters looked like on a given day. I simply "assume human control" of the team, make the transaction, then go on to the next team and transaction. I then revert my control back to the Braves, manage their game, then let AI sim the other games. I lock each player onto the AI teams' rosters.

I believe I set the rosters to 35-man. This does not matter, though, since the only team held to that limit is the team under human control. This is perfect, since ALL of a team's players imported onto their real teams during the assn setup phase.

My only real concern in this whole matter is that a lot of the rookies and seldom-used guys automatically import as 5-pointers (on the 1-100 scale). Since some of those 5-pointers are pitchers with STUFF=5, CONTROL=5, and VELOCITY=5, there are some ugly results.

[eddie01]

Well, I'm having fun with my replays, keeps me on my toes though! But, I would guess that looking in those never ending lines of code, would drive Shaun batty after a while. And finding the problem would be quite a challenge!

Unless I have a rather huge problem somewhere, I just run with it. Nothing in life is perfect![:)]

[Walewander]

Just thought you guys would get a kick out of this one. This is Brian Bannister, 2007, abilities around 40-25-40...

[Walewander]

Okay. I've started a new association. 2003 replay, player affinity on, ratings change off, injuries on... Just about every team has a pitcher with around 40 in stuff, 5 on velocity, and 40 on control. Walks and ERA are predictably ridiculous for these pitchers, and here's what the free agent pool looks like:

[puresimmer]

ORIGINAL: Walewander

Okay. I've started a new association. 2003 replay, player affinity on, ratings change off, injuries on... Just about every team has a pitcher with around 40 in stuff, 5 on velocity, and 40 on control. Walks and ERA are predictably ridiculous for these pitchers, and here's what the free agent pool looks like:

Did you dump all the players into the draft or is this a straight-up replay?