Send In The Clones! Nine Ichiro Suzukis vs. Nine Miguel Cabreras

[chucksabr]

Has anyone else here ever tried to do this kind of thing?

I'm a noob to the game, having just gotten it a little over a week ago, but there are already a bunch of things I want to do with it, e.g., create a fictional legend-level version of myself and have me start playing in my age 18 year; take over my favorite team the year after they have won a pennant (e.g., 1985 Tigers) and see whether I can keep them winning year after year; take over the 1963 Mets and see if I can get them to .500 and a pennant faster than in real life; etc. The possibilities have an end, but the end is a lot farther than you'd think.

One of the things I wanted to do was to play a team of a certain type of player versus a team of a very different type of player, and see which team wins more over thousands of games. The two players that leapt to mind were Ichiro Suzuki and Miguel Cabrera. Those guys are pretty different types of offensive forces, right? How would a team with nine Ichiros in the lineup fare against a team of nine Miggys in the lineup, everything else held equal?

The first question to solve is, how do you set it up? Here's what I did.

First thing I did was open up a 2013 Quickstart, dump all the players into the free agent pool, and eliminated 28 teams, leaving me with two.

I then imported Ichiro 2004 into the game. I went into player editor, went right to his defensive stats and made everything equal: all defensive ratings are 125, and all experience levels are 125. I changed his position to first base and renamed him Ichiro Suzuki 1B. Then I cloned him, changed the clone's position to second base and renamed him Ichiro Suzuki 2B. And so on, until I got eight defensive starter clones and a DH clone. Then I created four bench clones: backups at C, 2B, SS and CF.

I did the same thing with Miguel Cabrera 2008. The only difference was, when I changed the defensive ratings and experience levels to the same numbers Ichiro had, everything was exactly the same except one thing: Miggy's resulting rating at first base was 109 instead of Ichiro's 80. All other resulting ratings were the same. So I had to adjust Miggy's experience number at 1B to 91, to get the resulting rating to the same 80 Ichiro had. (Interestingly, I did not have to adjust Ichiro's RF experience to even him out with Miggy there. Go figure.)





Then I moved all the players/clones to the appropriate team—either the Flying Ichiros or the Slugging Miguels.

One of the things I noticed: even though I cloned the players all from the exact same guy, the batting ratings are not exactly the same for all the clones—nor, interestingly enough, are their star ratings. I find that very strange:



Can anyone help explain why this is?

Anyhow ...

Next, I had to create a pitching staff. I didn't think cloning these same guys and making them into pitchers was the right thing to do. So I did what I thought was the better thing: I created 12 pitchers (five SP, six MR and one CL), cloned each of them, and give each team one set each.





So now I had my teams. The next thing I had to do was set the lineups, which should have been easy: after all, the clones are all identical and field their positions identically, right? But I still did the due diligence and placed the appropriate Ichi and Miggy at the appropriate positions, with appropriate backups, just in cse I hadn't thought something through. I also did the same with the pitching staffs, making sure every setting was the same for both teams.



Lastly, I neutralized the strategy for both teams, running everything right down the middle for every split and every situation for both teams.

And then, finally, I ran the sim itself. I simmed 20,000 games under these parameters.

Result



The Slugging Miguels have a .506 winning percentage against the Flying Ichiros. What I don't know is whether this is staistically significant. Can any math genius here confirm?

The Miggys also hit four times as many homers. However, the runs scored is very close between the two teams, and the Ichiros' 19 point advantage in batting average and 10x advantage in stolen bases must have had something to do with keeping runs scored close.

Hitting



What leaps out at me here is not only how many more stolen bases the Ichiros have, but also what looks to be some 12,000 more triples than the Miggys. Of course, that doesn't make up the 25,000+ homer deficit the Ichiros have, but it does put a heck of a dent into it. As for the positions, those appear to have been randomly assigned before I went in and changed the depth chart.

Pitching



The only thing I really take away from this is that certain fictional players match up against one set of clones better than the other set of clones. Domingo Hernandez (2.14 ERA against the Ichiros, 5.50 against the Miggys) is a LH reliever with a 17 slider, 17 hold runners rating and 62% GB rate, but no changeup at all. On the other hand, Nelson Bonilla (3.45 against the Miggys, 5.86 against the Ichiros) is a RH reliever with a 20 slider, but holds runners only at a 4 level and has only a 3 changeup. So some explanations probably make sense here.

Top Performances - Batters




Not surprising that the Miggys have more top performances, since they hit way more homers, but they have 27 of the top 30 performances, which surprised me a little because of its dominance. Twenty thousand games, though, yet no four homer performances. Ichiro had a Rennie Stennett game, though. That's pretty cool to see.

Top Performances - Pitchers




This is really interesting: as much as the top hitting performances were tilted toward Miggy, the top pitching performances were tilted against Miggy: the Ichis had 25 of the top 30, including a no hitter and a perfect game against a team of Miggys. That does demonstrate that even simulation games occasionally punish slow-footed all-in-type sluggers occasionally, which makes the fact that the Tigers got shut out 11 times this year so far not so hard to understand.

So, the main thing I'd like to know is: has anyone else done this kind of simulation? And did I take it in the right direction, or is there something I messed up or didn't think through enough?

[Padreman]

Well one reason why the star rating will be different is if you forgot to adjust the fielding for each new position the clone will be playing, so low stats as an infielder will give him a loser star rating

[TGH-Adfabre]

One of the things I noticed: even though I cloned the players all from the same exact guy, their batting ratings are not exactly the same nor, interestingly enough, are their star ratings. I find that very strange:

Are you basing OVR on Positions or All Players? It is an option, I do not remember where. Some where in the League Set Up.

[chucksabr]

Originally Posted by Padreman

Well one reason why the star rating will be different is if you forgot to adjust the fielding for each new position the clone will be playing, so low stats as an infielder will give him a loser star rating

The numbers for fielding is exactly the same for every clone. I checked that specifically to make sure they were identical. What other setting could I have adjusted?

[chucksabr]

Originally Posted by TGH-Adfabre

Are you basing OVR on Positions or All Players? It is an option, I do not remember where. Some where in the League Set Up.

I just looked through League Setup and can't find what you're referring to?

[slugga27]

Originally Posted by chucksabr

I just looked through League Setup and can't find what you're referring to?

League setup>"Global" tab>bottom of the screen on the left-hand side.

[Padreman]

Originally Posted by chucksabr

The numbers for fielding is exactly the same for every clone. I checked that specifically to make sure they were identical. What other setting could I have adjusted?

Ok since Ichiro is an outfielder his fielding will be better on those positions but in the infield his fielding will translate low so it will have an affect on his over all rating

[Tyler87898]

I actually have done this type of thing where I make two extreme team of clones. I made a league with EVERY pitcher being the minimum ratings in the editor (except for stamina which is full). I made every batter have the best ratings for both teams. Also player defense was the worst possible and no fatigue for batters. It wasn't really like yours how your teams were different, I just wanted to see what kind of scores/stats would happen in this type of environment. Results? Players hitting 200+ homeruns with 40 WAR (what now Trout?) and scores being in the 100s. Interesting stuff.

[chucksabr]

Originally Posted by slugga27

League setup>"Global" tab>bottom of the screen on the left-hand side.

Found it. Now what do you think I should do?

[VanillaGorilla]

This is a difference of statistical significance. Assuming your .506 win% for Miggy meant that team won 10120 games, you are looking at a Z score of 1.67.

Conclusive? In a real life environment, real close to a yes. In a cyber environment, not really, because you can replicate it over and over and see if you get results that consistently favor the Miggy team, or not.

There is over a 95% chance that one team winning this many more games is not simply the result of "luck."

If you repeat the 20,000 games and the Miggies win 10100 games, or more, you are looking at more than a 99% chance of certainty that the Miggies, in your chosen environment, are better.

ADD: Just noticed the win total at 10122. This ups the Z score to 1.71 and gives you nearly a 96% degree of certainty that luck/randomness is not the determinant of your outcome. Same thing still holds as far as the 10100 figure giving you a 99% level of certainty, if you run the 20,000 game sequence, again.

[The Game]

Are you using 100% scouting accuracy? 125 RF should be less stars then a 125 LF a 125 2B should be more then a 125 SS

[pgjocki]

I find this report very interesting and might try and run a few simulations myself.

Consider, what if you had 1 Ichiro and 1 Miggy per team, how much does that change the equation or do they cancel each other out?

Or what if you cloned Kershaw for the pitching staff for both teams? How does Miggy fair against a left handed pitcher for every at bat?

Regardless, thanks for the good read and great idea.

[injury log]

Height is factored into 1B defense, which probably explains the discrepancy you observed in their ratings there.

I'd add that defensive ability is one of the differentiators between Ichiro and Miggy in real life, so neutralizing defense doesn't seem the way to go, in my opinion.

[chucksabr]

Originally Posted by injury log

Height is factored into 1B defense, which probably explains the discrepancy you observed in their ratings there.

I'd add that defensive ability is one of the differentiators between Ichiro and Miggy in real life, so neutralizing defense doesn't seem the way to go, in my opinion.

That may be, but the purpose was to compare their offensive contributions in as isolated a way as possible. If I could, I would have nine DHs in the batting order, and all neutral fielders and pitchers on the other side of the ball. I don't think I can with this game, though. Unless ... someone knows of a way to do that?

[chucksabr]

Originally Posted by pgjocki

I find this report very interesting and might try and run a few simulations myself.

Consider, what if you had 1 Ichiro and 1 Miggy per team, how much does that change the equation or do they cancel each other out?

Or what if you cloned Kershaw for the pitching staff for both teams? How does Miggy fair against a left handed pitcher for every at bat?

Regardless, thanks for the good read and great idea.

Thanks for the good words on the report.

Maybe I'm missing something, but I wouldn't think a single Ichiro and a single Miggy would have much more effect than the Mariners playing the Tigers in 2008 would have.

The pitcher idea might be interesting, if you could clone 12 of him for the staff versus 12 of another type of pitcher, using league neutral hitters and fielders to fill out the rest of the field.

[The Game]

Originally Posted by injury log

Height is factored into 1B defense, which probably explains the discrepancy you observed in their ratings there.

I'd add that defensive ability is one of the differentiators between Ichiro and Miggy in real life, so neutralizing defense doesn't seem the way to go, in my opinion.

Isn't this only for AI and real players that come in or have played 1B? My example of 4'10 Ewok Wicket won't work because I have him a s a bad 1B.

[TGH-Adfabre]

Originally Posted by chucksabr

Found it. Now what do you think I should do?

Against all players, I think, not know, it will even out some of the differences you see.

Fielding- Did you change their fielding attributes? INF Arm, Turn Double Play, OF Error, etc?

[chucksabr]

Originally Posted by TGH-Adfabre

Against all players, I think, not know, it will even out some of the differences you see.

I don't know what you mean by this. What exact setting do I need to adjust, and to what?

Originally Posted by TGH-Adfabre

Fielding- Did you change their fielding attributes? INF Arm, Turn Double Play, OF Error, etc?

I adjusted all the fielding attributes to 125 for all positions, all clones, both teams, to make it even. That was the intention, anyway.

[TGH-Adfabre]

I don't know what you mean by this. What exact setting do I need to adjust, and to what?

Sorry, I made no sense.

League setup>"Global" tab>bottom of the screen on the left-hand side.

I would select "Overall Ratings Based on All Players". I think it will adjust some of the differences in Stars you are seeing. That's it, it is a display issue for what you are doing.

[TGH-Adfabre]

I adjusted all the fielding attributes to 125 for all positions, all clones,
both teams, to make it even. That was the intention, anyway.

Ok, that is all I can think of right now. Cool test.