The File

[mike]

Why are good players in 100% of the other files, such garbage in this one (one coming to mind is Jose Reyes)? I am also noticing 22-23 peak ages.

Not every good player is garbvage by any means. Different sims create different results. It could have easily went the other way for Reyes.

[dejota]

24 is still early, IMO and according to the help file the average peak start should be 25, and with the slider setting I adjusted it should be 25.5. However, according to those numbers posted by Nick, this fear of people peaking at 22 and 23 is based, at least a little, in perception bias. It also seems those numbers are cumulative.

Also, 9 of the 10 22 yr olds who were peaked were drafted and played in college. Then once you get to the 23 and 24 year olds you start to see players that didn't make my radar for file edits or people I just plained missed an early peak start. I tried to catch as many as I could but over the course of 1500 edits you're bound to miss a few.

32 out of 2004 total players is only 1.6% of the population. To be clear I still feel this should be bumped to help minimize the number of draftees with a low peak. I think we need to decide if we want to bump avg peak start up a notch before we draft another batch of players.

[GoIrish]

I have had a few people peak at age 22. It is seams to be pretty random for the most part because I have had others peak at 25. It is part of the game in how random players peak, drop in rating, raise in rating.

For example on my team, I have had Rick Porcello and Julio Pena both peak at age 23. Porcello has dropped a ton in rating and Pena jumped up in rating. I have two hitters in my minor league John Carter and Jose Martinez at the age of 22 Carter peaked at 77 and Martinez at 22 is sitting at 72/80 still. Then I had players that have peaked already drop/raise in rating. I had Posey go from 83 to 85, Bucholz go from 90 to 92. I had Matsui drop from 81 to 79, Theriot drop from 81 to 78, Kubel from 81 to 79.

The basic gist is that it happens on all the teams. The whole part of the game is that you are not supposed to be able to predict the Roger Clemens from the Mark Priors or the Ichiros from the Fukudomes. That is why it is very random across the board. What fun would it be if every prospect reached their peak or you knew exactly what to expect from each player. The game is mimic of real life with baseball, think of all the one year flash in the pan players or the can't miss prospects that have missed. There are far more of them then the Derek Jeters or Alex Rodriguezs of baseball. It may frustrate us but what can you do, it is a game.

I think the biggest problem is what DJ alluded to earlier in his post. The drop in players and that in all honesty is on us. Think of all the players we have cut from our minor leagues that have gone to FA and then retired. That cuts into our player pool and that then skews the percentages because we are not able to replace those players faster than they are going out. We only have so many rounds in the draft and 1/2 those guys will get cut within a couple seasons when they don't pan out and then they will retire a season later after being in FA. I think this is why the drop in ratings for some have been so tremendous in some cases.

[AndyP]

I don't think anyone wants things to be completey predictable, but there is another extreme too - completely randomn. Realistically, things aren't entirely randomn in real life. Yes, it's impossible to tell for sure who the Jeters are at age 25, but you will at least have an idea about the quality, talent, and personality of the player such that you can take a pretty good guess. In real life there are tons of ways to make accurate predictions about player success and in the ways Mogul tries to quantify these you can be very good evaluating talent if you know what to look for and how to balance factors. This file, unlike any I've ever experienced, appears to be incredibly randomn. In other words....too far to the randomn side of the spectrum. Yeah, things like Markakis and his 92 injury rating being out for 3 months is going to happen - I get that. But the randomn injury drops, vital drops, vital stall, etc. are what I'm concerned about.

If you were a major league scout and at the end of the day your evaluation came down to a roll of the dice rather than predicatable, observable factors - you'd probably say "screw going to watch a game, I'll just roll the dice". That's a bit how I feel right now about this file and Mogul 2011 - in an effort to get away from being too predicatable it has become too randomn. I guess I personally would prefer too predictable. Too randomn eliminates the element of skill.

I'm not saying DJ did anything wrong, I'm starting to think 2011 may be a problem that the designers are refusing to acknowledge and he didn't have a prayer putting together a good file. It's simply frustrating how much fluctuation the talent seems to be subjected to.

[GoIrish]

I don't think anyone wants things to be completey predictable, but there is another extreme too - completely randomn. Realistically, things aren't entirely randomn in real life. Yes, it's impossible to tell for sure who the Jeters are at age 25, but you will at least have an idea about the quality, talent, and personality of the player such that you can take a pretty good guess. In real life there are tons of ways to make accurate predictions about player success and in the ways Mogul tries to quantify these you can be very good evaluating talent if you know what to look for and how to balance factors. This file, unlike any I've ever experienced, appears to be incredibly randomn. In other words....too far to the randomn side of the spectrum. Yeah, things like Markakis and his 92 injury rating being out for 3 months is going to happen - I get that. But the randomn injury drops, vital drops, vital stall, etc. are what I'm concerned about.

If you were a major league scout and at the end of the day your evaluation came down to a roll of the dice rather than predicatable, observable factors - you'd probably say "screw going to watch a game, I'll just roll the dice". That's a bit how I feel right now about this file and Mogul 2011 - in an effort to get away from being too predicatable it has become too randomn. I guess I personally would prefer too predictable. Too randomn eliminates the element of skill.

I'm not saying DJ did anything wrong, I'm starting to think 2011 may be a problem that the designers are refusing to acknowledge and he didn't have a prayer putting together a good file. It's simply frustrating how much fluctuation the talent seems to be subjected to.

Andy,

i think the reason it is so random and volatile is because of the loss of players. It has skewed our percentages that are built into the game. We can sim forward but the sim forward will not account for the human element, which is us cutting the prospects that are not panning out and after they sit in FA, then opt to retire. I think we have to assume that our plateau of players might be around 2k, if that is the case, maybe we should alter the percentages to match what we were going to see with 3k players.

[AndyP]

The other problem, and this is completely seperate from the league, but I don't think that the designers of Mogul are attacking the right elements for randomness. Production and injuries are randomn/unpredictable. Not skills. Mogul has seemingly taken the approach of randomnly dropping or adding skills rather than having players have up and down years. Vitals need not necessarily follow production, that would better capture the realistic challenge of scouting.

---

Andy,

i think the reason it is so random and volatile is because of the loss of players. It has skewed our percentages that are built into the game. We can sim forward but the sim forward will not account for the human element, which is us cutting the prospects that are not panning out and after they sit in FA, then opt to retire. I think we have to assume that our plateau of players might be around 2k, if that is the case, maybe we should alter the percentages to match what we were going to see with 3k players.

Yup, I saw DJ's post about that. I think Mogul should look at adding more depth to the draft if the total population has wild affects on talent. But I agree, we should play with the settings to make sure that our talent levels remain somewhat stable despite population boons/dips.

I was more commenting on a mogul problem in general too. The designer has flat out said he was aiming for a randomn effect and I wonder how much control we'll have even tweaking our file to stop those adjustements he made.

[GoIrish]

The other problem, and this is completely seperate from the league, but I don't think that the designers of Mogul are attacking the right elements for randomness. Production and injuries are randomn/unpredictable. Not skills. Mogul has seemingly taken the approach of randomnly dropping or adding skills rather than having players have up and down years. Vitals need not necessarily follow production, that would better capture the realistic challenge of scouting.

Without a doubt that is true, the one thing I don't like about the AI of the game is random drops on players that are being productive. You would think that would be the opposite.

[AndyP]

The other problem, and this is completely seperate from the league, but I don't think that the designers of Mogul are attacking the right elements for randomness. Production and injuries are randomn/unpredictable. Not skills. Mogul has seemingly taken the approach of randomnly dropping or adding skills rather than having players have up and down years. Vitals need not necessarily follow production, that would better capture the realistic challenge of scouting.

Without a doubt that is true, the one thing I don't like about the AI of the game is random drops on players that are being productive. You would think that would be the opposite.

Yeah, there are multiple threads on the mogul site about that - the designers and main posters there seem to be pretty defensive about this issue because they like the randomn nature of the drops.

But like I said, the problem is that what they have randomnized is completely unrealistic. Frankie Liriano was still throwing 94-95 last year with the same slider and change he has this year, but for non-quantifiable reasons he had a down year. This year - he's back. It's not like his "vitals" dove, he just didn't pitch to his talent. THAT is what they are trying to simulate, but they're doing it the wrong way. Randomize production, don't randomize skill adjustments.

I think the changes they made to create this silly form of randomness may make file tweaking almost impossible. At least that's my worry.

[dejota]

Well I think our best bet, since we can't change the coding for mogul, is to try and identify where the most fluctuation is coming from. To play devil's advocate for a second, Clay was attempting to mimmick players up and down years by increasing the volatility of their stats. This system would be more effective and less random if the population wasn't also volatile.

I truly believe we have hit that population plateau, as Sean referred to it. When I simmed ahead it seemed the population trended upwards by about 100-200 year which would be evened out by humans being more apt to release a crappy minor leaguer.

I'm also going to play w/ the draft class settings (increasing peak start/% of college players since they're a source of frustration/etc.) and see if I can get it to generate one w/out a bunch of low peak starts consistently. Once I do that I can generate another class (so I'm not privy to info you guys aren't) and that should at least stop the infusion of players that will peak out at or before our league average of 24.

The best we can do is monitor this class (and compare it to the previous 2 classes) and see if this is an effective approach. I know the IFA's I make will not have that early of a peak start, so if we closely maintain a flow of talent that's not going to peak out terribly early we can attribute the fluctuations more to the random nature of talent development. (I also pointed out that I mistakenly moved the player development slider the wrong way which is only making their career path's more random, not by much but it's still notable).

I think we these three small measures we can maintain the intent of the new features of 2011 (since we can't avoid them) and nudge the talent towards a more, but not completely, stable nature. I'll diligently notate what I change in the file tonight and the results it produces. Aside from this I don't know what else to do or say about this topic.

[mike]

So after tonights sim will teh draft class thats on the file be the one we are working with? Just wondering because I'll likley scout the draft tonight.