Normalization: A Humble Request to Strat-O-Matic

Moderator: Palmtana

  • Author
  • Message
Offline

egvrich

  • Posts: 1436
  • Joined: Thu Aug 30, 2012 4:17 pm

Re: Normalization: A Humble Request to Strat-O-Matic

PostThu Aug 18, 2022 9:15 am

STEVE F wrote:I sent this entire thread to SOM support. This was there response:

Hi Steve,



There is no truth to that whatsoever - there is no normalization going on that affects players or teams based on what they happen to be on pace for. There are no hidden mechanisms designed to achieve any sort of unexplained agenda. There's not even a reason we can see for us to do anything like that.



Sincerely,
Strat-O-Matic 365
https://365.strat-o-matic.com


Steve

That is the same basic response I got two days ago to an email I sent them over a month ago ... I'm guessing your email prompted the response that I received:

There is no truth to the accusation that there is anything going on in our game engine besides what we have always outlined publicly and officially. There is no normalization going on that affects players or teams based on what they happen to be on pace for. There are no hidden mechanisms designed to achieve any sort of unexplained agenda. There's not even a reason we can see for us to do anything like that.


Sincerely,
Strat-O-Matic 365
https://365.strat-o-matic.com
Offline

Backfire

  • Posts: 208
  • Joined: Sun May 17, 2020 7:20 pm

Re: Normalization: A Humble Request to Strat-O-Matic

PostThu Aug 18, 2022 10:09 am

FrankieT wrote:Backfire,

Welcome back as it seems you returned from a hiatus. I took a break from ATG V-VII so I get it. But in your posts a few things stand out that may be worth considering and have broad applicability to many discussions here.

Use of the term "Normalization".
As djmacb pointed out, the initial assertion that SOM admitted to something was taken out of context. Normalization is not what you (and others in this forum BTW) describe. It has a specific meaning that makes your logic hard to follow because it is not related. In general, normalizing data is about putting disparate data sets onto a "level playing field" for comparison or synthesis, if we are using bad puns. It is not related to what you assert, although maybe you are just using it nominatively. But in that case, choose a different name for clarity.

Selective anecdotes.
For instance, another common point of view that is used as evidence of something nefarious is when players have something like a "sharp decline" in HR output. But we never hear about players who have a sharp increase in HR output--though I know it occurs--we all experience it. And HR could be anything--hits, walks, etc. It is a selective anecdote that leaves out the rest of the story, which invalidates the extreme extrapolation from a point-in-time anecdote to a massive corporate intention. It is quite an unsupported leap statistically and logically.

Subjective expectations of card behavior.
This last one is also common. A misunderstanding of statistics while claiming to use statistical bases for assertions. Player cards not conforming to someone's expectations has many underlying assumptions that are not typically true in the cases on this forum.
It assumes that the long term expected statistical outcomes are proven or disproven by short term non-valid sample results, or subjective assessment. Think of it this way.
If you run a league and have a team, maybe you win 85 games.
If you run that same scenario another 1000 times, you may have an instance of winning only 70 games, and maybe an instance of winning 90. Maybe your 85 wins was a low probability outcome and it over-achieved. That is, a result that is in the tail of distributed outcomes. It does not mean that your team would be expected to win 85 games every time it was used--but that might be your expectation and source of frustration when that team wins 70, which may even be closer to expectation.
The same applies to an individual card. Which BTW--the best way to gauge expectation is to look at the actual results of a card under the same conditions multiple times. There is simply no other way to do it because of the varying factors of the playing environment such as opponents, salary cap, stadia, etc.

Overall, are there differences in the online version from cards and dice? Yes.
Are there differences between the online game and PC/Home version? Yes.
Are those differences posted? Yes. See the wiki (is it 100% updated? I don't know, but that's a different debate).
Are there some legitimate calls for increased transparency of results, that are NOT related to nefarious intent by SOM? Sure. But to conflate a call for display of all PbP results and lookups etc (as I would advocate for) as being equivalent to SOM purposefully altering results to favor new players? This is not supported by any facts. Zero. Especially the most vexing one which is that new players generally do poorly.

Just a few thoughts. Welcome back.


I should perhaps not refer to it as normalization in that case. Point taken. Regarding selective anecdotes, anomalies spring up in data constantly and in this case, the sample sizes are minute and thus are of course not adequate proof of anything. However, I strongly suspect there is more going on than we realize. Unfortunately, I have absolutely no way to prove anything, nor do any other players as none of us have access to the code. I may be the outlier here but we are oddly in a very similar position.

I do understand how averages work, especially highs and lows. What you said is true but there is also intuition and that is something which cannot be quantified and is easily dismissed. It would be nice to have all the answers of how the mind works, and how the game works, but we sadly do not. I have learned to trust my intuition.

I am not sure why you would think new players would not do poorly in a pure environment. Do you think the level of skill expression in the game is low? In practically any form of competition, new players tend to do very poorly. This game is very demanding. There are thousands of player cards, a plethora of ballparks, quite a few rules, many of which are not easy to pin down, and a very tough field of players which is dense in veterans. Sure, the game is largely about dice rolls but team construction and managerial decisions and settings should pose quite a challenge for any new player.

I have been playing games my entire life and this is the only one I do not have easy access to the rules. So yes, added transparency, perhaps complete transparency would go a long way to dispel my concerns. As Dr. Biocide said, we just want to know the rules. Speaking of which, I recently played against a team with atrocious hold ratings and decided to set my base stealing team to very aggressive. I usually do not use this setting but this seemed like a prudent time to try. I was caught quite a few times and I wish I knew the reason. Yes, this is perhaps just another anomaly, just bad luck. But I suspect that setting is overturned and probably a trap. Knowing how it works would be great.

Thank you for your reply and welcome, it's nice to be treated like a human being and not a lunatic out of touch with reality. Especially since I know firsthand that game companies 100% do have secret variables, controls, etc that players are never privy to. That is a fact in the industry and not one I enjoy but it was established long before I signed up. So to you guys, it may seem excessively paranoid or unrealistic but in the eyes of many game designers, it is considered necessary. You might be surprised at times how deep the rabbithole truly is.

Having said that, I don't necessarily think the code was designed with ill-intent, nor do I think Strat-O-Matic is doing anything "nefarious", but I do think they may have concluded that controls are needed to preserve the realism of the game. Having 150 win teams and earth shattering homerun records that could veer so far from real baseball stats may be offputting to some players.
Offline

Bballexec

  • Posts: 42
  • Joined: Tue Mar 31, 2020 12:27 am

Re: Normalization: A Humble Request to Strat-O-Matic

PostFri Aug 19, 2022 3:25 am

Backfire,

Generally, I find that people do not appreciate what random means. In Strat, what is Random is the sequence of events. During, any given at bat, the outcome will be something from either the hitter card or the pitcher card. Some events on those cards have a very low chance of occurrence, but since they are on the cards, they will occur at some point. In Strat anything that can happen, will happen given enough dice rolls.

For example, in my newest 2021 league, Kittredge my closer gave up 12 run ER in 8.67 IP thru the first 21 games. On his card, he gives up a HR on 6-5, with a split between 1-6. That roll came up twice with a HR result. He also gave up a BP HR on 4-12, with the park factor of 4. This resulted in 3 BS and losses for the team. These numbers may not come up again for the rest of the season. Do you think his ERA should be over 12 for the rest of the season? By game 45 it is down to 3.99. That is not the computer normalizing in some way. It is just different dice rolls coming into play. I have Kittredge in another league and his ERA is 1.62 thru 105 games. He has probably been a bit lucky in the that league, although he is helped by a pitchers park and a very good defense.

In your base stealing example above, you only focused on the pitcher hold ratings. You did not mention the catchers arm rating, or the runners ability to get a lead. Even if those things were favorable as well, it is still possible to have a lot of runners thrown out in a series. Such is the nature of random numbers.

When I started playing two years ago, I ordered a five team game set from Strat to get a printed copy of the rules and all the charts. That was interesting, but it convinced me I did not want to play the super advanced game manually. I am quite happy to let HAL handle all the calculations needed to play a game.

In my view, the $80 million salary cap does serve to level the playing field to the point that I have wondered whether all the results are random. I have only played 41 teams and have a pretty good record so they probably aren't totally random. However, I probably put a lot more work into each team than other players, so it is hard for me to have more than 2-3 teams going at any one time.

There are times when I think I have the better offense, defense and pitchers and still lose games. That happens in real baseball as well. You look for small edges and try to build good teams for the selected ballpark, but some times you don't get the dice rolls.
Offline

Coachbobg

  • Posts: 4
  • Joined: Mon Apr 20, 2020 11:39 am

Re: Normalization: A Humble Request to Strat-O-Matic

PostFri Aug 19, 2022 12:19 pm

I think playing around to see what works and what doesn't work is part of the fun. Sometimes the results are quite surprising but to me that's all part of it.
Offline

Backfire

  • Posts: 208
  • Joined: Sun May 17, 2020 7:20 pm

Re: Normalization: A Humble Request to Strat-O-Matic

PostFri Aug 19, 2022 2:46 pm

@Bballexec I hope my posts don't come across as whining, I should be clear that I dislike it when I feel I am being favored as well. For example, I oddly had over 300k left on a team and couldn't find a place to spend it so I put it into James Heath as a backup. He entered the game when an injury occurred and I immediately thought "Here we go." And of course, he hits two home runs and a triple. Then next game he hits another triple. The sad part is that this is just an anecdote and one could say "Yep, it's random." Anything I point out except for raw data and a large sample size can be quickly dismissed. However, I believe in my intuition and I also know that nobody here wants anything to be rigged, so we should all have some measure of bias against such claims. I thought about posting examples but it is completely pointless. All I'll say is the mind is a fascinating thing, we don't fully understand its scope. It is very easy to discredit intuition, feelings, etc but that doesn't necessarily mean they're wrong. In fact, I would go so far as to say that there are many occurrences in life that are not "random" or what we may describe as normal. I'm sure you've experience some very strange, seemingly impossible things in your life. But I digress! :)
Offline

labratory

  • Posts: 428
  • Joined: Sat Sep 29, 2012 11:33 am

Re: Normalization: A Humble Request to Strat-O-Matic

PostFri Aug 19, 2022 8:04 pm

Humans are absolute shit at pattern recognition and are always seeing patterns where none exist. You know who thinks they're really good at pattern recognition? Gamblers. Yet somehow the house always wins. Perceiving and tolerating chaos is the more valuable skill to cultivate.

I always feel sorry for the guy that drops half a paycheck on the Detroit Lions because he hears that "the Lions are 8-1 in September when playing as a home underdog against a left handed quarterback".

On the other hand, my teams are 1-7 when facing a $0.50 starting pitcher. I hope nobody reads this and starts Luis Leal against me. haha
Offline

Hack Wilson

  • Posts: 1133
  • Joined: Thu Aug 23, 2012 6:16 pm

Re: Normalization: A Humble Request to Strat-O-Matic

PostFri Aug 19, 2022 8:06 pm

I agree with what FrankieT says, he explains it very well. I've been playing SOM online since 2001, and the game has improved and evolved. Could it be better? Yes. Do I expect perfection? No. Would I like everything to be transparent? Yes. I don't expect it, nor require that, though. It's just a game.
Offline

FrankieT

  • Posts: 1313
  • Joined: Sat Mar 03, 2018 12:07 am
  • Location: Usually Somewhere Else

Re: Normalization: A Humble Request to Strat-O-Matic

PostFri Aug 19, 2022 10:26 pm

These two statements actually struck me as an alpha and omega of sorts.

lab..."Humans are absolute shit at pattern recognition and are always seeing patterns where none exist."

Hack..."It's just a game."

Indeed
Not trying to rouse any ire--just unexpectedly complimenting an interesting intersection of thoughts...from not just two successful managers, but a couple guys I like to "play" this game with because I know they will compete to the fullest and never throw their hands up in frustration.

It is why if I have a 108 win team that gets swept in the semis, I tip my hat, offer congrats and good luck, and move on to the next challenge.
Offline

labratory

  • Posts: 428
  • Joined: Sat Sep 29, 2012 11:33 am

Re: Normalization: A Humble Request to Strat-O-Matic

PostFri Aug 19, 2022 10:44 pm

lab..."Humans are absolute shit at pattern recognition and are always seeing patterns where none exist."


Actually, this quote originally came from MAX power. I agree with MAX but wanted to make sure that the original author was credited.

As an aside, how do you get the quote to say:
"FrankieT wrote:" at the beginning of a quote? I thought it happened automatically but apparently not.
Offline

FrankieT

  • Posts: 1313
  • Joined: Sat Mar 03, 2018 12:07 am
  • Location: Usually Somewhere Else

Re: Normalization: A Humble Request to Strat-O-Matic

PostSat Aug 20, 2022 7:11 am

labratory wrote:
lab..."Humans are absolute shit at pattern recognition and are always seeing patterns where none exist."


Actually, this quote originally came from MAX power. I agree with MAX but wanted to make sure that the original author was credited.

As an aside, how do you get the quote to say:
"FrankieT wrote:" at the beginning of a quote? I thought it happened automatically but apparently not.

Ah--thanks for the proper hat tip...Hat tip Max. And of course he is also among those successful managers that are fun to try to beat/compete against.

For quote, I just hit the quote button on your reply.
Or, if the post is very long, can do it manually with typed or pasted sections as
Code: Select all
[quote="username"] TEXT [/quote]

which yields
username wrote: TEXT
PreviousNext

Return to Strat-O-Matic Baseball: All-Time Greats

Who is online

Users browsing this forum: No registered users and 11 guests