LINK: 10 Lessons about Projections

by

From ZiPS inventor Dan Symborski, over at The Hardball Times, his 10 Lessons about making baseball projections. All very interesting and at the same time they feel quite familiar to me, echoing many of my experiences.

Except No. 8, which talks about the regression of inseason stats, which is not as pronounced as from season to season. This means that a player who gets off to a hot start for half a season, is more likely to continue that hot start in the second half than he is to have it continue the next year.

That doesn’t surprise me for any particular individual, but in aggregate it does.

This 10 Lessons story is part of 10 Lessons week at THT. Mitchel Lichtman’s piece about defensive statistics is full of great stuff, too.

 

 

How’m I Doing? The End of April Report

by

I’m playing in three leagues this year. Tout Wars NL, American Dream League AL, XFL Mixed. This is about as close as I’ll get to the ideal of putting all one’s eggs in one basket, where attention is focused and mistakes hurt all season long.

So, how are things going?

TOUT WARS NL

The standings are bleak, and they’re not getting any better. Well, I did climb into 10th place last night, thanks to Todd’s nightmare.

Screenshot 2014-04-30 11.54.42

You can see all the league details here.

I went with an extreme Stars and Scrubs strategy here, and a pitching staff of Bumgarner, Rafael Soriano and a bunch of cheap guys. The hitting has not been helped by Ryan Zimmerman’s injury, and as expected picking up productive hitters via FAAB has not been easy. Tony Campana has helped with four steals. Otherwise, not so much from anyone.

The good news is that Nate McLouth should see more playing time due to Bryce Harper’s surgery, and I have an insane cushion in OBP. The question coming out of the auction was whether I would be able to deal OBP for productive hitters in other categories. It’s still too early to judge, but if I’m going to climb out of this hole it will be because someone saw lots of value in Joey Votto.

The other good news is that the pitching staff has been pretty good, despite their pathetic standing in WHIP. Cheap guys out of the draft, Wily Peralta and Tanner Roark, have been good. So has cheap FAAB pickup Alfredo Simon. Bumgarner’s WHIP has been a big problem, as has Edwin Jackson’s performance overall, though that has improved lately. There’s work to be done here, but if Jake Arrieta isn’t bad and Andrew Heaney shows up in June, there is some potential to be a pretty good staff.

Getting off to a bad start is a problem. It makes it more difficult to maneuver, and puts pressure on that leads to mistakes. My week by week finishes (11, 10, 10, 3) show some improvement. There is still lots of time, if I make the right moves.

AMERICAN DREAM LEAGUE

I’ve been in first or second all season thus far.

Screenshot 2014-04-30 12.10.51

I came out of the auction thinking I had a pretty strong team, both offensively and in pitching. Until last week I was languishing in homers, however, but then Kyle Seager busted out, and things are okay there.

Where I am suffering right now is stolen bases, despite having guys like Alex Rios, Shin Soo Choo, Erick Aybar, and Ian Kinsler. And wasn’t Eric Hosmer supposed to run a little, too? The problem is that all my speed guys are pushing into their thirties, so it shouldn’t be a surprise they’ve slowed down some. I’m going to need to do something about that.

Especially because I didn’t buy a closer, despite intentions to. I did buy Matt Thornton, as a CIW, and scored, sort of. He’s saved three games, which is better than nothing, but obviously if I don’t add steals I’m going to need to trade some speed for saves. It’s going to take a lot of luck to improve in both categories.

Last week was a rough one, because I lost Chris Sale and Shin Soo Choo for extended period, and dropped into second place. Shoo came back last night, and Sale is expected next week, probably. This team has broad enough talent that I should be able to compete, but it’s still early. Plenty more could go wrong.

XFL

This odd mixed league, with an auction in November and a 17-round reserve draft in March, is a keeper league. Alex Patton and I are co-owners.

Right now we have a strong offense, but have been hobbled by our big three starters, Jordan Zimmerman, Jered Weaver and Mike Minor. We also have a team construction problem. Since these guys aren’t big strikeout guys, and we’re playing three closers, we can’t really compete in strikeouts.

This being a mixed league (15 teams), the success of the secondary starters is very important and we’ve not gotten much out of Jenry Mejia, Felix Doubront, Jarred Cosart and some others. Henderson Alvarez has been excellent, but it’s hard to expect him to continue on this level.

This is a league in which teams play to win, and when that is clearly not in the cards they trade for next year’s keepers. We’re in poor enough shape to have started to think about the next step, but we have so much talent it’s hard to embrace failure so soon. Alex is chomping to make changes, I’m keeping my fingers crossed. But losing Archie Bradley is not a harbinger.

Screenshot 2014-04-30 12.25.56

ASK ROTOMAN: Holliday for Machado? Blame It On Cain.

by

Hey Rotoman:

I have Longoria as my 3B, and have Manny Machado. I want to trade Machado to improve my outfield.

What do you think of this trade: I get Matt Holliday for Machado and Matt Cain?

“Machado Man”

Dear Macho:

Some music while we think about this.

First off, I don’t have enough information to judge. It sounds like you won’t miss Machado much, playing Longoria instead, but whose working week does Holliday end? And who is filling Cain’s shoes?

The problem here is that Cain has not been a Top-20 starter so far this year, and that makes it easy to undervalue him after a few bad outings. I recently l0oked at last year’s slow starting pitchers and found that pitchers who were expected to be good usually did okay after a bad start. Matt Cain was one of those guys last year, and he recovered, at least part way. A bad start doesn’t mean the rest of the season is a disaster.

Still, over the five months after his bad start Cain wasn’t the Cain anyone who bought him last year hoped for him to be. And it has to be disturbing that he’s starting this season in similar fashion. So I can understand disenchantment.

Okay, another musical interlude.

But I can’t really judge your trade because I’m not seeing all the pieces. The two parts that stand out are: It’s a great idea to upgrade your outfield by dealing a reserve corner. Not that it isn’t valuable to have a solid backup like Machado, but to win you probably have to maximize even if that’s a little risky. So that makes sense if Holliday is a nice bump on the guy you’re replacing.

But Matt Cain shouldn’t be a throw in. If you have a great staff without him, making him expendable because of his poor start, I don’t have a problem with that. But if you’re dealing him because you’re expecting him to be replacement level the rest of the way, I think there’s a fair chance you’re selling him a little short.

Ably,
Rotoman

ASK ROTOMAN: Stanton for Braun?

by

Rotoman:

I play in a Head to Head points league, with -1 for strikeouts and -1 for errors.

My friend requested a trade where  he gives me Giancarlo Stanton and I give him Ryan Braun. I was about to hit yes when I decided to look at the stats and Giancarlo strikes out an awful lot more than Braun. He also commits more errors. Should I go for the boom-or-bust Stanton, or keep my very solid Braun.

“First I Look At the Stats”

Screenshot 2014-04-25 18.25.57

Dear First:

Good idea to look at the stats. Here’s some back-of-the-envelope projections:

Braun should score a lot more runs (+30), hit roughly the same number of homers and drive in a lot more runs (+20). That’s the benefit of playing on a better team in a better stadium. Braun will have a higher batting average, and, oh, he will run more than Stanton, too.

Braun also has an advantage in strikeouts (125 to 140 for Stanton) and errors (6 vs. 8), but these differences are small beer.

On Stanton’s side, he’s still young and is likely to have some years in which he hits more homers than Braun ever did. But it’s hard to see that happening as long as he plays in Miami, so that’s not relevant this year.

Also on Stanton’s side, Braun is older, which usually leads to an increase in strikeouts at some point. But Braun isn’t worrisome old yet at all, so if you’re looking for an edge there you’re reaching.

Obviously, it’s your decision, but I’d hang onto Braun.

Conclusively,
Rotoman

 

ASK ROTOMAN: How Much Is Potential?

by

Dear Rotoman:

For several years i have been waiting for two players to become regulars, hit around .280, and become true value players. No. 1 is Joaquin Arias (SF Giants) and Luis Valbuena (Chicago Cubs). I still think both have good potential, but they are not so young as they were a while ago. Do they ever reach their potential as i envision it?

“Two Prospects (Past Due)”

Dear TPPD:

It seems like a long time ago I had a bit of a thing for Luis Valbuena. I look back now on his stats and I’m mystified. I paid money for this guy? He has always been terrible at the major league level.

But he wasn’t always playing at the major league level. In Triple-A his stats were pretty good, year after year.

So, first of all, don’t rely on Triple-A stats for major league projections. Good major league candidates often skip Triple-A. And plenty of players who are pretty good but not major league ready thrive in Triple-A. We call them Quad-A, FWIW.

Luis Valbuena is one of those guys. Get over him. The clock has ticked and you were wrong about him. He can’t hit or run or hit for power. Over.

I never much cared about Joaquin Arias. Nice name, but he hasn’t nicked my platinum prospect shine.

Maybe that’s because as an offensive player he’s been offensive. He has had some defensive value, if you like Fangraph’s WAR evaluations. But even if you don’t, there is no way you can possibly say that Joaquin Arias is a breakout offensive candidate this year or in the future.

Potential is the ability to outperform expectations, which tend to settle in the middle. Luis Valbuena and Joaquin Arias may have better times, they could play better or they could be used better, but so far they are near to the dregs. At their age and previous performance, it would be lunacy to think they have any future potential.

And I think they serve as a good cautionary tale about putting too much faith in Triple-A stats and defensive prowess.  A good fantasy strategy is to avoid losers. Luis and Joaquin are doing fine in the real world, but in the fantasy world they’re losers.

Sincerely,
Rotoman

LINK: What Makes A Fan?

by

Screenshot 2014-04-22 23.12.43This NY Times story digs deep into Facebook Like data to discover what teams Facebookers are fans of, then maps that to the Facebooker’s ages when their teams won World Series.

When I was eight the Yankees lost the World Series to the Cardinals, then spent the next 15 years in CBS and Steinbrenner hell. It was hard not to like Horace Clark, but it was in those years when I was nine to 17 that my love affair with the Mets took hold.

Over at rockremnants.com I think we see constant evidence that the bands we fell in love with when we were 15-25 or so are the ones that stick with us forever. Baseball seems to ring a somewhat earlier chord.

 

ASK ROTOMAN: The Rich Get Richer

by

Rotoman:

6×6, 8-team head-to-head categories league.  I just got Jose Reyes back from the DL, so who should I drop from the following:

Jose Altuve ($23)
Chase Utley ($17)
Dee Gordon ($7)
George Springer ($10)

Thanks.
“Reyes of Light”

Dear RoL:

Oh, the agony of the shallow league. I’m not going to address 6×6 and Head to Head issues, since you don’t specify categories and H2H mostly ups the randomness of winning and losing rather than adding any strategic nuance.

The numbers in the parentheses above are what those players were going for in preseason auctions. Jose Reyes, by the way, went for $25 or so. So, the obvious answer would be to drop Gordon., but I can understand your reluctance. He’s already stolen 10 bases and while he isn’t going to continue to hit .365, if he hits .270 on the year he’s going to steal a lot of bases.

Jose Altuve is off to a solid start, making him one of the best second basemen in the game, so it’s hard to drop him, and Chase Utley is off to a monster start and has more power than Altuve, Gordon and Reyes, so you have to hold onto him.

Which leaves Springer as the odd man out. He was the surprise promotion last week, and there is a lot of excitement about him because he has a nice power/speed combo. There are also serious questions about his contact rate, and he’s struck out five times in 15 plate appearances already—on track with his Triple-A numbers, which could prove a problem.

One of the agonies of playing in a shallow league is deciding between excellent or exciting players. If Springer goes on to hit 15 homers this year, you may look back and wish you’d kept him rather than Dee Gordon, but at this point Springer’s risk of failure is high enough I think you have to go with the speedy middle infielder.

Sincerely,
Rotoman

 

ASK ROTOMAN: What is FAAB Worth?

by

Dear Rotoman:

I was thinking about redeeming the DLed Nate Jones in Tout Wars AL, which got me thinking about FAAB, Vickrey bidding and being in last place a few weeks after the season starts. I wrote about it here at USA Today.

Does it make sense to add Jones’ $14 now? Or should I wait to see if he can get healthy and reclaim the closer job in Chicago?

“Ron Shandler”

Dear Ron,

I had some thoughts about your recent column and thought they made a better topic here than at your various discussion boards. Though perhaps I’ll show up at those, too.

You raise a few interesting issues that I wanted to touch on.

I will admit that I thought the Tout Wars redemption process was going to be a disaster. It has instead been a great success. (I thought the same about FAAB trading as well, and was wrong about that, too.) In Tout Wars teams are allowed to cut any player on the DL and reclaim their draft day price as FAAB up until the All Star break—after which they can reclaim half their draft day price as FAAB.

The  usual dynamic for redemption is determining how long a player will be out, versus his utility when/if he returns, filtered by the value of the added FAAB. This means that guys who are out for a while, but who may not be out all season, can reside on a team’s unlimited DL reserve until it becomes clear that the extra FAAB is going to matter (as we approach the break, and the interleague trading deadline).

What makes Nate Jones interesting is that he could return to the bullpen eventually and not gain the closing job, which would pretty much waste his $14. That’s a good reason to cut him and reclaim his bid price, though at this point he’s not close to returning, and it is pretty unlikely that the $14 you add to your FAAB total will have any utility at all until much later in the season.

For those reasons, I suggest waiting until he’s either done for the year, close to returning as a non-closer, or you need the $14 to buy something better.

As for Vickrey, the bidding auction system that we use in Tout Wars, it awards the FAAB player to the highest bidder, but reduces their cost to $1 more than the second highest bidder. Cory Schwartz was quoted earlier this week about how he thinks Vickrey just randomizes the process, and he doesn’t like it, but his bidding last week in Tout Mixed Auction is a prime example of Vickrey’s importance and why you and I like it.

As we all know, there were some closers available in last week’s bidding. Cory decided he needed one of them. Zach Steinhorn agreed with Cory that the best available closer was Francisco Rodriguez. Zach bid an aggressive $33, but Cory trumped him by bidding $60, which was then reduced a la Vickrey to $34. One can look at this as Cory “saving” $26, but it is a fairer evaluation of Vickrey to say that Cory bid aggressively because he wanted K-Rod most. Such overbids are made knowing that someone else who did the same thing would raise the price of K-Rod a lot, but that was a price Cory was willing to pay. The stated intention of Vickrey auctions are to limit system rigging, since bidders are encouraged to bid the absolute most they’re willing to pay (knowing that if they value more than the market they won’t have to pay their full price). Cory bid what he was willing to pay, and since no one else would pay as much, he ended up with a discount, as it were. That’s a feature, not a bug.

Where Vickrey excels is when there are a number of bidders. Where Vickrey falters is on the players for whom there is a limited market. With only 12 or 15 teams in a league, many without holes at particular positions, there may only be one or two teams looking for a player at a particular position. There may only be one or two of those players at that position available. One of those teams may value one of those players a lot, but chances are, even if he bids aggressively, his bid will be reduced to $1 or a few dollars because there was no market for that player. That seems to me to distort the bidding process.

For a couple years we played in Tout Wars with a $10 floor on bidding. If you bid $10 or less, that was the price you would pay if you won, with no reductions. If you bid $10 or more and no one else bid more than $10, your winning bid would be reduced to $10. If two bids exceeded $10 the standard Vickrey rules applied. The idea was to increase the cost of roster churning at the low end, where the market is less than robust. Many objected to this, saying they thought that if we were going to play with Vickrey we should play pure Vickrey. After a rule change, that’s the way we play now, and while I still think it makes the low-level bidding somewhat arbitrary, it isn’t really a problem.

I recommend Vickrey bidding for the most contested players, but the use of a floor for the cheap bidding. That’s the best balance in my opinion.

Okay, back to FAAB and inseason values. One rule that might help us find the balance between Draft Day dollars and FAAB dollars would be to combine the two. Let’s say teams are given $360 on auction day, and are told that they can spend as much of that as they like, with the balance ending up as their available FAAB balance. How much would they actually spend on Draft Day?

Or, less radically, you’re restricted to the $260 for your regular team, but then can bid FAAB $ for the reserve rounds.

By increasing the porousness between Draft Dollars and FAAB budgets, we open up ways for teams to play different strategies at the draft table, in the reserve rounds and all season long during waivers and claims.

Earlier this year I looked at how many stats were available via FAAB and claims in the NL and Mixed Leagues. This is what our money goes to buy on draft day versus what we’re able to add as the season progresses. Would that number change if we spent more cash on draft day and had less available for inseason buys? It sure looks to me that paying more on draft day is the way to go.

Sincerely,
Rotoman

 

 

ASK ROTOMAN: Time To Give Up On Salazar?

by

Rotoman:

Is it worth giving up Danny Salazar in a 5×5 keeper league where I need some power in order to get Matt Adams?  I already have Madison Bumgarner, Adam Wainwright, Robbie Erlin, Yordano Ventura, Tyson Ross, and Jeff Samardzjia for SP’s plus a couple others. What do you think?  The other offer was Bumgarner for Mark Trumbo, but I think thats too spendy.

“A Bridge Too Salazar?”

Dear Bridge,

Let’s get the obvious out of the way. You are pretty rich in pitching and need a power hitter. On draft day Salazar and Adams probably cost a similar amount, and it sounds like in hindsight if you could have taken Adams instead of Salazar you would have. So this deal is obviously to the good for you.

And I agree about Bumgarner and Trumbo. The former was a $24-25 player in NL leagues, while Trumbo was a $22-23 player. If that was your only way to add power you might do it, but it isn’t as square a deal as Salazar for Adams.

It might have a bigger impact, since Trumbo is a more significant hitter than Adams, and Bumgarner is a more significant pitcher than Salazar, but sometimes it’s better to play it safer and smaller, which is what Salazar for Adams does.

That’s because we know quite a bit about Adams. He’s a very solid power hitter against right-handed pitchers (and not so good against lefties). He may lose some at bats versus lefties at some point, especially given the Cardinals’ plethora of hitters in the bigs and the minors, but he’s been weak enough against lefties (.588 OPS in limited time the last three seasons) that you probably want to see him sit against them. He’s still a solid power hitter even when missing those at bats.

We don’t know quite a bit about Salazar. He took the AL by storm late last year, in 52 IP, and was quickly elevated to stud status  by many despite the lack of experience. I warned about that during the preseason, because regardless of a young pitcher’s obvious gifts (he throws very hard, missed lots of bats with good control, etc.), we simply don’t know how well and quickly he’s going to be able to adjust as hitters adjust to him and his workload increases. Salazar has not been good thus far this year. His velocity is down, he’s walking guys, and throwing homers. It’s early, just three starts, but things are not right right now.

So make the trade. Getting out from under the struggling Salazar is really just a slim advantage. Odds are Salazar is going to work things out, and once he’s in a groove he’ll throw plenty of strikeouts, but he’s not in that place now, and if you can fix your team by dealing a struggling cipher, you have to do it.

Enthusiastically,
Rotoman