Thurston Fluff

Well played sir. Well played.

GM's say a lot of things that are not accurate. Being a PR man is part of the job. PR sometimes involves spin. Not ot mention gathering talent is the GM's job so he's got several million good reasons to say the talent level is higher than the production level. To say some teams do better than they should has a built is assumption that the predictive models of what they should do to begin with are by definition accurate. I disagree with that. I believe what actually happens to be more accurate. As the KC example I used earlier shows those predictions are sometimes wrong. You say the predictions are right about the talent level but KC just got lucky? Two years and 324 games in a row? I say the talent assessment model got it wrong and there was far more talent than it said there was. Yet I am silly for using wins while you are not because you used stats that failed to make an accurate prediction? To assume any as of yet incomplete system is by definition right and realty got it wrong is what sounds silly to me. Fair point. But if 100% accuracy to your barometer every way to assess talent is equally silly to use. I'd also argue the Seattle example is a bit misleading if you look at the whole story. In the off season they hired new general manager, field manager and an entirely new major-league coaching staff not to mention scouting staff changes. When assessing a team's talent support staff has to be part of it. If, for instance, the old regime was terrible at scouting upcoming opponents, didn't use proper defensive alignments, mismanaged players or any number of other factors involving non-players the players true value may have been miscalculated. In essence you are assuming the models had it right and their talent level was not what their record said it was without taking into consideration the support staff might have had something to do with the players assumed talent level. That said, it was a huge swing and some of that I think had to do with attitude and other non-talent factors. So I guess you are accurate to a point. You just left out some qualifying factors that helps explain why it was such a huge swing. If the 08 team was on the low end of projections and the 09 team the high end it gets us half way there. Adding in excitement for a new regime probably played some role in it. Yet the other aspect that is talent related is the support staff was totally different. Major staff changes, like player changes, effects the team's overall team talent level. The next part is less clear to me. You said they improved their talent and their record went down. There are two ways of looking at that. They added actual talent and they still got worse or they added what they thought was more talent but it turned out not to be so. Given how poorly Zduriencik was at assembling talent I don't think it's silly to say it could have been a misjudgment of talent. If talent is only 50% of the reason teams win then any system that predicts wins based on talent would also only be 50% accurate. Do you really believe that? Oh and health is part of team talent. Depth is part of talent. The better depth, thus overall talent, mitigates health issues. Not to mention the ability to prevent health issues, while not nearly 100, is at least partially a matter of the people who acquire, use and condition the players. Cal Eldred was overused by Garner and he got a very predictable injury because if it. That is part of the managers talent level. Or more accurately lack of it. Again if non players are part of team talent, and I think they obliviously are, then their ability has to be rated as well. But honestly I don't know where you get the notion that 50% of the wins is completely out of the team's control. They aren't forced to forfeit half their games. It isn't predetermined a team has to lose that day? Correct me if I'm wrong but I think you are saying something to the effect the best of teams are bound to lose a certain number of games and the worst teams are bound to win a certain number of games. Something to the effect that every team wins 50 games and every team loses 50 games. If that is what you meant it's a complete misinterpretation on your part. That is not the same as saying the games are out of a team's control. It's simply saying there are enough variables to allow lesser talent to win any given game. Over the course of the year there will be a certain number of games where the best team does not win. That is not the same as half the games are out of a team's control. Furthermore if you played two teams 162 times, the better team would come out of top almost, if not every, time. Obviously if they are almost identical other factors play a larger role thus the need for the ?almost" qualifier. If there is any reasonable difference in talent the best team will always come out on top after 162 games. I'd go even further to say the greater the disparity between the two team's talent level would be closely reflected in the disparity in wins. I agree we should be talking in ranges rather than exact numbers. I'd put it at about a five game swing either way but I can get on board with a 10 game swing. I do not agree that there is a wide distribution outside of that range in any given year. While there can be outliers I disagree that it is at all common. I'll concede three points though. 1- wins is not 100% accurate. There are, and will continue to be, outliers. 2- There is a range of outcomes so a team's record is not in perfect harmony with talent level. On the flip side I think it's fair to say all other ways of assessing talent are not 100% accurate either. My belief is they are much less so. You, and obviously a lot of other people, say more so. 3- things like attitude, having something to play for, or a team not using it's best players because of a lost season can be a factor. Though I am not sure if talent isn't part of attitude and such. Personality and motivation are kind of murky areas when it comes to skill set IMHO.

Strawman. No one is saying that. But if you actually think that's what everyone else is saying, I guess I don't know where else to go with this. You are more than free to believe what you like. You just accused me of using a strawman argument by using a strawman argument. I was using a hypothetical scenario to show what you were implying by using that implication within another context. Thus the word imagine. That certainly does not imply I thought someone thinks that way. I was hoping to show how close the relationship between winning and talent is. None of the other factors comes close to it. Even combined they don't. If talent isn't more important to winning then why would anybody go through a rebuild? Just keep a 500 talent team on the field and wait for the other factors to kick in. Teams are willing to go through painful rebuilds because it has so much more effect on wins than anything else it doesn't pay to sit at .500 and wait for all the other factors to kick in. To sum up 1-My basic argument is more talent causes more winning. (I doubt anyone disagrees with this.) 2- I also say it is by far the largest cause. So much so that all the other factors combined do not come close to determining how many wins a team will get. (This is where I appear to be at odds with most people here.) My conclusion based on those two premises is, 3- If the single most important factor to winning is talent then the number of wins best tells you how much talent you had. Now if you disagree with my assertion that talent is not that much more important than luck, a team playing above their collective heads, a perfect storm of events or whatever, fine. If you believe the conclusion based on my premises is wrong please show me why. Just don't try to make it out like I am making strawmen to build my case.

Health is part of talent level. Deeper teams compensate better for injuries than thin ones. Same goes for trades. While the Royals projections couldn't account for Cueto coming in it took enough organizational talent to acquire him That is part of overall team talent. Luck is certainly part of it but i don't buy into it playing a major factor simply because of the number of games played. Over 162 games things like luck evens out. IT might acount for a few games but that is about all. SO if you want to say a team had 68 win talent instead of 70 game talent I guess that's fine. The unbalanced schedule is fair game. Which is why I was always against it. But if that was a major factor it's hard to see how three teams in the toughest division in baseball had the three best record in the league. I just don't buy into luck playing a significant role in how PECOTA missed by 18 games a year after they missed on the same team the previous season. Especially since they had everyone coming back and went to the playoffs the year before. It's even more telling that it is harder to get from average to good than it is to go from really bad to slightly below average. They missed terribly. I'm guessing PECOTA also believed their system was correct and thought it was pure luck that made them miss so badly. Which is why they didn't look to change their computations when they missed so badly. Which of course made them miss again. Got to make you wonder how long it will take them to figure out it's them not the team that got it wrong. With the exception of the unbalanced schedule and luck all those variables are part of the talent level. I believe luck plays a very small role for reasons I stated above. The unbalanced schedule does play a role but again I don't think it's a huge one. Imagine a GM saying talent isn't all that necessary to winning. There are multiple way to win without having a lot of talent. We can just wait for the schedule to favor us or get lucky. Maybe we'll gather players who do well in wet or cold weather and wait for a year when we have an unusually high number of wet or cold games. Teams gather as much talent as they can for a reason. Talent is the primary dictator of wins. Thus it stands to reason the number of wins is the primary indicator of a team's talent level.

To get it as wrong as it was cannot be explained by good health and playing over their heads.

If the best one says something the others don't why would we go with the lesser ones? For example when a batting average says the hitter is good but more accurate stats say he isn't you don't give batting average the same weight as the other ones. Same here. Even that is besides the point. The point of all of the things we try to measure is to predict a future outcome. Once you know the outcome, if it doesn't match the estimates you don't throw out the result and say the estimates are correct anyway. You look at why they were wrong and try to improve on them so you can more accurately predict future events. Sometimes I think people view analytics as a completed venture. They are not. They are ever evolving. This is not 2+2=4. They are a hypothesis that says A, B and C factors should produce X,Y and Z outcomes. When the results don't match the prediction you can't say the results are wrong because your hypothesis didn't predict it. You say the hypothesis needs tweaking. I'm not trying to be a wienie about this but I think sometimes people get so involved in what should happen they forget the tools that say what should happen is not a completed study. We need to question why something didn't work in reality like it said it should on paper. Otherwise we'll never get closer to both things matching like they should. Saying reality got it wrong seems to me the wrong approach to getting there.

I ordered some stuff form one of our vendors this morning. Then I had to do another order this afternoon. As I was doing the second order the person I was on the phone with said the person I ordered from this morning needed to ask me something before I got off the phone. When she got on the phone she asked me about the morning order and I answered it. I was curious as to how she knew it was me on the phone so I asked. Apparently they know it's me if the person can't stop laughing. Not sure what to think about that.

Why is it necessary to have a "best" indicator when there is lots of available evidence from which to draw conclusions? You're really asking why it's necessary to use the best indicator? For the same reason we don't use batting average and RBI over OBP and slugging as an indicator of hitting ability. One is better than the other.

I guess my point is what best determines talent level if not the record? PECOTA who determined the Royals had 72 win talent last year? Or one of the other ones that similarly rated the Royals as around a 500 team the past couple seasons? Does anyone really think the Royals just really outperformed their predicted record two years in a row?

Actually I think that is exactly what it means. So the most talented team in baseball always wins the World Series? Or at least the 10 most talented teams in baseball always make the postseason? Winning 3 series is not at all the same as what happens over 50 series. I'm quite sure you know that so I don't understand what your point is there. As far as the most talented teams always getting to the playoffs that is accurate. The actual measurement used is who wins the most games. If that is the test then the ones who win the test are the best. The talent test is not which team has the most players who's stats best fit a predictive model. It's who wins the most games. I think there can be time when a team who is almost as talented as another can pull the upset. When two teams are close to each other then things like the unbalanced schedule (something I wish they would change) and dumb luck play a larger role. Now if his point was a team who wins 68 games is really only a 65 win (or something to that effect) level fine. What doesn't happen is a team with low talent suddenly wins a boatload of games or vice versa. Injuries aside of course. The overall idea that a team's talent is, or should, be based on anything but how they actually do is simply wrong. The reason we know certain talents are beneficial is because of how many wins they provide not the other way around. If a team does not provide the wins projected then the talent wasn't what was predicted.

Actually I think that is exactly what it means.

It is an extreme oversimplification to say since Stearns came from the Astros he will do exactly what the Astros did. It completely ignores these franchises are in totally different conditions. We have a top ten farm system. The Astros at the same point in their rebuild had one of the worst. The Astros had almost no major league or upper minor league talent at the same point in the rebuild. The Brewers have both. The Astros were more like the Brewers when Dean Taylor took over than they are the Brewers of today. If Stearns cannot see there are differences between the two team' situations and act accordingly we hired the wrong man. Personally I think we hired the right guy.

Yes, and there were some pretty good players on that team. Lucroy will be a huge factor in if they lose 100, pretty much every position is average or below average offensively and defensively. I have watched a lot of Brewers baseball and looking at this team it is right up there with the worst Brewers teams no doubt. The future is much brighter but it doesn't change the current major league team and how horrible they will be. In 1994 Rickie Bones was our lone all star. Though it was a strike shortened year they would have had to go 9-35 the rest of the way to be a 100 loss team. In 1995 Kevin Seitzer was our lone allstar yet that team scraped together 65 wins in only 144 games. It is very hard to be a 100 loss team.

You seem really fixated on the Astros way but seem to think the entire plan was to tank to get top draft picks. Yet tanking has pretty little to do with either of those team's success so far. This article has a pretty nice breakdown of how much the top picks from tanking has helped get them to where they are now. http://espn.go.com/blog/sweetspot/post/_/id/68059/dont-credit-tanking-for-the-success-of-cubs-and-astros It's how the Astros scouted and developed their talent. Take the year that the Rookie of the Year Carlos Correa was drafted. They got him to sign under slot to give them more to sign their 2nd round pick. Lance McCullers Jr. Both were integral in the Astros getting into the playoffs....and will play a huge role in years to come. Every team knows drafting and developing is the way to go. That is not a secret. How to do it is. My beleif is the Astros success is not based on tanking and getting the top pick. It was by being better at finding and developing players. Virtually all of which did not involve tanking to the top draft pick.

You seem really fixated on the Astros way but seem to think the entire plan was to tank to get top draft picks. Yet tanking has pretty little to do with either of those team's success so far. This article has a pretty nice breakdown of how much the top picks from tanking has helped get them to where they are now. http://espn.go.com/blog/sweetspot/post/_/id/68059/dont-credit-tanking-for-the-success-of-cubs-and-astros

There is also a risk that a good deal never gets made because he is waiting for the ideal trade to come along. The odds that Davis is going to get us more than one of the better catching prospects in the game isn't likely to happen.

Nice trade all around. Davis seems like an ideal DH candidate for a small market team hoping to contend and the Brewers get a catcher who should make the majors and a pitcher who could. Even better if neither of them have to be on the 40 man roster for a while. Flexibility and talent, what more can you ask for?

I'm ok with missing on injuries or the occasional oddball season of a small number of players. What I would like to know is how accurate are they overall in projecting things. I'd really like to know how close a sabre oriented projection of players gets to reality. Especially since our team is going to use more of this sort of thing to get us to the promised land.

Umphry, thanks for taking the time to explain things. It helps understand the situation better. Good info but does anyone know how accurate they have been? I get the luck factor can create outliers but if it is worth anything, even as a baseline, it has to get us in the ball park for the average player. (Pun intended) How well has it predicted things in the past?

Are there that many other places looking for a older bench player with fringe major league talent, no upside and little history of success at the major league level?

Yet Gennett gets an base at a higher rate than Carter. Plate discipline is a good thing but so is the ability to actually hit the ball. Obviously Gennett's approach has served him better than Carter's when it comes to getting one base. I might add that Carter's extra patience at the plate hasn't really helped him hit the ball much better either. Carter has a career OPS of .764 vs Gennett's .742. I just don't see how Carter's patience makes him better than Gennetts swing away approach. That said I think Gennett would be better served to take a few more pitches. It could only help his numbers. Not sure how useful of a comparison that is. Scooter would not be anywhere near as good if he wasn't platooned most of the time. Also, Carter in a down year still had a higher OBP last year than Scooter. Gennett also plays 2nd which makes lower offensive thresholds more acceptable. My point was not to compare the two as much as it was to say just taking a few more walks doesn't make him a better hitter.

Yet Gennett gets an base at a higher rate than Carter. Plate discipline is a good thing but so is the ability to actually hit the ball. Obviously Gennett's approach has served him better than Carter's when it comes to getting one base. I might add that Carter's extra patience at the plate hasn't really helped him hit the ball much better either. Carter has a career OPS of .764 vs Gennett's .742. I just don't see how Carter's patience makes him better than Gennetts swing away approach. That said I think Gennett would be better served to take a few more pitches. It could only help his numbers.

Keep in mind Gennett is younger than Walsh. In the time he was playing in the majors Walsh was mostly in A and AA. He has a total of a 147 ab's above AA. He OPSed .703 over those ab's. I get it's nice to see what they new guy has to offer but lets be realistic about his upside. He certainly isn't likely to out hit Gennett this season. Nor will he get us anywhere near the trade Gennett can bring us if he can hit right handers like he generally does. Walsh should not be getting starts over a younger player who has shown he can actually hit above the AA level several years ago just because he's the shiny new toy. I get he had a great season in AA last year. However, as guys like Hunter Morris show, any player can have a good season and still not be very good. Just like a 25 year old who has a tough year in the majors doesn't necessarily mean he's permanently regressed. It was worth taking a shot on Walsh due to the team's situation. If he makes it as a major league player all the better. But to take over as a starter from someone younger than he is who has already shown he can actually hit major league pitching in his still young career? No thank you.

Not to nit pick but it's been done here before. It's how we got the Braun, Fielder era.

http://bleacherreport.com/articles/2611725-undefeated-high-school-wrestler-lets-opponent-with-special-needs-win Makes me proud to be a wrestler.