I have nothing to quote, but WIS staff have said multiple times that no performance modifiers are built into the SIM, besides the known 4.5% platoon advantage. Normalization does modify a player's probabilities, but in direct relation to actual league year stats, not based on any arbitrary increase or decrease to the underlying statistics.

I would recommend sending in a support ticket with the question if you are unconvinced

I appreciate the responses- enlightening info and helpful suggestions.

Posted by uncleal on 6/19/2017 3:22:00 PM (view original):

Posted by jmcraven74 on 6/19/2017 2:58:00 PM (view original):
Thank you guys for the input, but to go back to a question I asked that I don't think was addressed- how do we know to a certainty that no drag or balloon effects are applied to a player?

For example, many "battle" games apply a +/- 5% randomization to a units base statistics in matchups, has WIS ever said they do not this?

They have said they do not do this for multiple games. They have never commented on an individual game.

However, it seems likely that over 162 games, all those individual game adjustments should come close enough to canceling out that we don't need to worry about them much.

The more I look at what goes on in the leagues here, the more this just doesn't pass the smell test. I mean this toward WIS, not uncleal. I think there's something they're not shooting straight with us about.

Like uncleal mentions, over 162 games, it should all wash out. Besides a + or - bonus tagged on a player, I can't see how a player with weak normalized stats can have a WIS best season playing on the same team with a player with significantly better normalized stats across the board who has a worse season than the weaker player. Same parks played in home & away, same pitchers faced. Over 162 games, the randomized individual matchups should balance out, but there's infinite examples of when they don't.

Is it possible that the random matchups produce a really bad season from start to finish for a player with great normalized stats? Sure. But is possible that it happens with the frequency and extremity we see regularly while, with equal frequency and extremity, we see random sub-par normalized guys on the same team have a breakout year? Personally, I don't think so. It happens with every team I draft--some players overperform all season while others underperform. And I think the dice are re-rolled when the playoffs start.

That wasn't what uncleal said. He said that, if single game variance exists (which we have no reason to believe it does), positive and negative values for it should even out. That is NOT the same as saying that there's no reason a worse player should outperform a better player. Just because the better player has a 10% better chance (at most, in most cases) of getting on base every time for 600 times, that does not mean that 600 times is necessarily a large enough sample to produce that result 100% of the time (or even close to it).

Posted by ozomatli on 8/11/2017 2:27:00 PM (view original):
That wasn't what uncleal said. He said that, if single game variance exists (which we have no reason to believe it does), positive and negative values for it should even out. That is NOT the same as saying that there's no reason a worse player should outperform a better player. Just because the better player has a 10% better chance (at most, in most cases) of getting on base every time for 600 times, that does not mean that 600 times is necessarily a large enough sample to produce that result 100% of the time (or even close to it).

Correct. Even a whole season is, by statistical standards, a small sample.

To clarify, I don't believe that single-game variance exists either. I acknowledge the possibility that it does, but don't care, because it would even out.

Not to mention the possibility of park effects, pitcher usage quality, etc. changing results...

A good team will have better offensive numbers... this sounds like a truism, but there's a bit more here... teams that win by a lot of runs a bit more will see more 200K mop-ups, allowing hitters to pad their stats...

I understood what uncleal said, and I didn't intend to misrepresent it. My apologies if it came across that way. I simply meant that, in my opinion, if single game variances would even out over a full season (162 instances) so too should a player's normalized performance even out over 700 instances.

As far as a good team having better offensive numbers because they will see mops, or park effects-- that's not really going to have much, if any, impact on what I'm talking about, which is the difference in normalization deviation among players on the same team.

What I'm talking about is, within a fairly controlled environment (the same team) a player by player evaluation of the season-long deviation between regular season performance and normalized stats.

And I don't mean that I expect every player to hit his normalized stats every season (obviously there are factors that will impact that)--what I mean is I think it's odd when I see so many instances of inferior Player A having a +25% deviation from his normalized stats while superior Player B has a -25% deviation from his normalized stats while they are both playing on the same team, obviously playing in the same parks against the same pitchers. And when I say superior Player B, I mean he is equal to or better in each normalized stats category-

1B/100#

2B/100#

3B/100#

HR/100#

K/100#

BB/100#

AVG#

OBP#

SLG#

OPS#

If I understand you correctly, it sounds like you're saying 162 instances of full-game variations (a full season) is a large enough sample size that "it would even out", but 700 instances of plate appearances (a full season) is too small a sample for normalization deviation to even out.

And honestly, in my mind, the question should not be 'can you possibly find a way to explain away every instance of significant conflicting normalization deviation among players on the same team?' The question should be 'why are there SO MANY instances that require an explanation?'

Posted by jmcraven74 on 8/13/2017 1:04:00 AM (view original):
I understood what uncleal said, and I didn't intend to misrepresent it. My apologies if it came across that way. I simply meant that, in my opinion, if single game variances would even out over a full season (162 instances) so too should a player's normalized performance even out over 700 instances.

This is an incorrect assumption.

Can we please stop perpetuating the myth that WIS somehow knowingly and intentionally applies artificial statistical (dis)advantages to individual players in individual seasons (or even single games, really?) or for specific owners only. The range of possible outcomes over a relatively tiny statistical sample of 162 games or 700 PA is enormous and individual SIM seasons will seemingly always produce outliers or strange results, especially if you're looking for those results as "proof" that these artificial manipulations exist. To achieve any kind of evening out of statistical performance to an expected mean for an individual player would take far, far more than 700 PA.

WIS hasn't yet followed through on the 2012 SLB redesign and hasn't made any real significant engine updates since forever ago (was the fielding update in 2009?). Hard to believe they're somehow manipulating individual player performance during all this time.

One problem we also have is that the engine hasn't been updated in so long that, to borrow from card game terminology, "the format/metagame is solved". We already know exactly who the best buys, most consistent performers are, etc. and in OL or OL-like leagues, those see the most play. That's why so much of the userbase doesn't play OLs anymore -- it's boring.

I actually think dynamic pricing was WIS's way of "unsolving" the game without having to do real work (i.e. update the engine). Changing salaries changes price efficiency and people start looking for new players.

Should be REQUIRED reading for new players, and even a lot of experienced ones...

Should I spend significantly more on a couple star hitters or equally on all of them?

I need help in general. I've made the playoffs maybe two times and both that I remember started out with 2-0 leads and lost the series 3-2.

I think I rely on RC27 too much and my team struggled on pitching.

There's no one right answer on how to build your team. From glancing quickly at your teams, you're not too far off target. You could probably get away with drafting fewer innings, and probably might benefit from a few more PA from your main guys. As far as whether to spend on high dollar players or split it more evenly, it can work either way, but look for value. A lot of the really good players have had their salaries raised so much that they aren't worth it anymore, so I personally find better luck with a more evenly divided team.