Feet to the Fire
Ah, Projection Season. One of the most glorious times of the year. Each year, I find myself getting more and more excited as each player projection approaches, barely able to contain my enthusiasm. Oh, wonder of wonders, I say! Oh, heavenly endeavour! Oh, what joy unbounded, what pure and glorious-
Wait a second. What the hell am I talking about? I hate projection season.
It's true. I know, I know. I write for a stat-heavy baseball blog; how can I possibly not like projections? Well, let me tell you, it isn't easy, but I find a way to make it work. I almost never take part in these projection excercises, and while I certainly find it interesting to look at the various projections, I just don't find it very enjoyable.
More than anything, though, projection season is an ugly, in between time of year. Most of the good hot stove stuff is already done, as the good free agents are pretty much all gone (this year being somewhat of an exception), and you don't see too many trades until closer to spring training. You're tired of talking about last season, having analysed pretty much every single iota of information possible from several hundred different angles. And worst of all, it's still a full month before pitchers and catchers report. Let's face it, folks. The middle of January is baseball purgatory.
But, in the spirit of trying to be more projection-friendly, I've decided to do something that several of the posters around these here parts have suggested: I'm going to look at our past projections and see how we did. See, I figure that nostalgia is one of my chief allies in the world of the written word, so I shall gird myself with the weapons of the past as I attempt to find pleasure in numbers.
Now, it has been suggested that someone should round up all the projection systems, compare them to the actual performance of the players, and then try to determine how accurate these systems really are. Well, aside from the fact that I'm sure someone much smarter than me over at BP or the like has probably done so at some point, I also simply don't have that kind of time (nor, to be perfectly honest, ambition), to take on such a massive project. No, I thought it just might be fun to take a look back at our very own VEB projections from the year most recently ended, and see how we did.
Last year, the VEB community did projections for six players; four pitchers and two hitters. Honestly, I was surprised that there were so few, but I went through the archives, and that's all that I saw. If I missed one somehow, I apologise. I'm going to stick to the pitchers today, both because I'm strapped for time, and also because frankly, the pitching ones are really, really interesting.
Slight side note: you should all really read through the original threads on all of these; it's fascinating to see all the people who don't come around here anymore. It makes me wonder why.
Player Projection #1: Kyle Lohse
| GS | IP | H | BB | SO | HR | W | L | ERA | WHIP |
| 29 | 184 | 196 | 56 | 118 | 21 | 11 | 11 | 4.42 | 1.371 |
| 33 | 200 | 211 | 49 | 119 | 18 | 15 | 6 | 3.78 | 1.300 |
The top line is the VEB projection, the bottom line is Kyle's actual numbers from last year.
So, how did we do? Well, pretty good, actually. What jumps out at you, of course, is the ERA; somehow, Lohse managed an ERA about three quarters of a run better than what we expected. The culprit? Lower home run and walks totals. Of course, this isn't really news, as we're all well aware that those two things were the real keys to Lohse's success last year. He gave up hits at about the rate that we expected, and the strikeouts are fairly comparable. He didn't walk many, and he kept the ball in the park. Simple, huh?
Player Projection #2: Joel Pineiro
| GS | IP | H | BB | SO | HR | W | L | ERA | WHIP |
| 27 | 169 | 181 | 51 | 110 | 21 | 11 | 10 | 4.44 | 1.377 |
| 25 | 148.2 | 180 | 35 | 81 | 22 | 7 | 7 | 5.15 | 1.446 |
So, how did we do? Not quite as good on this one, honestly. Seems that at least a few people around here were looking for the Jo-El who handcuffed the Mets to show up, as opposed to the one that actually did. Looking at the numbers, it's easy to see why El Pinata was so much worse than we thought: he gave up what is commonly referred to as a 'shit ton' of hits, and he didnt strike out much of anybody. You want the definition of a hittable pitcher? Look no further. He also gave up homers at a faster pace than we predicted; not a good combo when you're already giving up a lot of hits and allowing contact on pretty much every pitch you toss up there.
Player Projection #3: Anthony Reyes
| GS | IP | H | BB | SO | HR | W | L | ERA | WHIP |
| 24 | 147 | 141 | 52 | 114 | -- | 10 | 9 | 4.43 | 1.313 |
| 6 | 49 | 47 | 15 | 25 | 4 | 4 | 2 | 2.76 | 1.265 |
So, how did we do? Do you have any idea how much I love this one? Not because we learned anything at all from it, but just because it gives me a chance to think about ol' Iron Bill and the organisational differences that led to his departure again without feeling like I'm just self flagellating for no reason at all.
To be honest, this one isn't even close to being right, but in our defense, none of us thought that Anthony's 2008 was going to be anything like it was. At the time this projection was made, he was seen as being the favourite to be the fifth starter in the rotation; we all know, of course, that did not, in fact, happen. All of his starts were after he was dealt to the Cleveland Indians; Reyes served the Cardinals only in a relief capacity.
The one and only thing that really jumps out to me about this (and it is significant), is that in both projection land and reality, Anthony gave up less hits than innings pitched. That, we were all kinds of accurate on. In fact, the rate at which he gave up hits in 2008 is exactly the rate we predicted. Let me say that again. Our projection on Anthony's hit rate was perfect. I'm not a huge math guy, so maybe that's not as weird as I think it is, but estimating a rate stat like that perfectly strikes me as more than a little bit freaky.
This was also the only projection that lacked a home run component for the pitchers; not sure why, exactly, but there it is. The long ball, of course, always was a bit of a bogey for Anthony, and I'm assuming that our predictions for his ERA were based largely on a continuance of his homer-allowing, no-runner-left-behind anti stranding policy ways. What happened for Anthony is that he mostly kept the ball in the yard, reducing his HR rate to less than one per nine innings pitched, and kept his walk rate down fairly low, less than three per nine innings. In short, the only thing Anthony really didn't do that you would like to see is strike people out. In the minors, Reyes consistently had remarkable K/BB rates, and as we all know, those ratios are one of the best predictors of future success we have. Since coming to the big leagues, though, his strikeout rates have been nowhere near what they had been. Bottom line, I still believe in Anthony Reyes. He's maybe not quite what we thought he was going to be, but that's part him and part a bad situation, in my ever so humble opinion. The talent is still in there, I think, and I hope Reyes figures out a way to tap into it fully in Cleveland.
Player Projection #4: Matt Clement
| GS | IP | H | BB | SO | HR | W | L | ERA | WHIP |
| 24 | 142 | 141 | 60 | 112 | 17 | 9 | 9 | 4.48 | 1.421 |
| 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
So, how did we do? Well, I saved the best one for last. At the very least, it's certainly the cleanest, tidiest pitching line you're ever going to see.
I'm going to go on record right now and say that I really liked the Matt Clement signing at the time. I did. I may look stupid now, but at the time, I thought it was an excellent, calculated risk. Of all the bargain guys on the market that the Cardinals were legitimately looking at to try and catch lightning in a bottle for the back of the rotation, I thought Clement had easily the best stuff in the non-Mark Prior category, and should have had at least a decent chance to come back healthy.
Sadly, it wasn't to be for Clement. His velocity just never came back, and it looks like it very well may not be coming back at this point. You feel bad for the guy, who really was a pretty good pitcher at one point. He got lost, of course, in all of the wringing of hands and gnashing of teeth over the Mulder and Carpenter arm issues and the Wainwright injury. (Good god, how did this team manage to win any games at all last year?)
Given what we know now, I would support a Matt Clement signing again. Tomorrow. Seriously. I don't mean Clement specifically, of course, but this type of pitcher. Sure, I understand that it feels like we're always dumpster diving, and it can be frustrating at times. But even so, I don't see any real downside to taking a shot on a guy like this. Worst case scenario is exactly what we got out of Matt Clement last year: nothing.
Well, I hope you've enjoyed this trip down pitcher projection memory lane. I'll probably throw the two hitter projections in another post as part of something else one of these days, but I'll be honest: neither one is nearly as compelling as a couple of the pitchers.
Take care, everybody. And Mo, if you're reading this, I'm just kidding about Clement. Seriously, no more of that stuff. You guys are killing me.
Oh, one last thing. If everyone could be so kind, I would greatly appreciate if you would head on over to the RFT's interview with Tony La Russa. He's promoting his ARF music thing, so if you've ever wondered the skipper's exact postion on Dragonforce, this is your chance to find out. It isn't one of my pieces, but a) it's always a fun time when the skipper starts waxing about his favourite rock bands, and b) it would score me major points with my editor if I can steer some traffic over. So come on, El Vivis, show the Baron how much you love him. Do if for Red. Red would do if for you.
(This moment of third person speechifying has been brought to you by Rickey. Congrats, Rickey.)
61 comments
|
0 recs |
Do you like this story?
Comments
I would support a Matt Clement signing again.
Ask and you may receive
Roch Kubatko talked to starter Kris Benson, who said his shoulder hasn’t felt this good since 2000. Benson had rotator cuff surgery almost two years ago. Benson says the Cardinals and Dodgers have expressed the most interest, with the D’Backs and Rangers also in the mix.
Four Clubs Interested In Kris Benson
Stat Whore
Khaaaaaannnnnnn!!!!!!
It was half my fault, and half the atmosphere.
by the red baron on Jan 14, 2009 12:13 PM EST up reply actions
I so miss the Juan/Khan picture
probably the biggest reason why I miss Encarnacion
* sarcasm might be involved in this comment
Khaaaaaannnnnnnn!!!!!!
double effect, like the movie.
Cardwash - Cardinal, Washington fan (Washington???? Yeah, I know)
Saw that on...
MLBTR this morning, and found the Clement-like signing statement very ironic. I still think the signing was worthwhile…it just didn’t work out. Maybe this time we’ll make it a point to make sure Benson is throwing well before we go and give him $1mil.
How come there’s no common character for million? K and G = 1,000…we need to establish something for million so I don’t have to type out mil anymore. We could go with M for million or Miles (about what $1mil should by you), but that seems to generic.
Sometimes I wonder,
"Why is that frisbee getting bigger?"
...and then it hits me!!
by cardzfanbub on Jan 14, 2009 12:28 PM EST up reply actions
I think M *is* the abbreviation for million
youneverknow
by floodOfLove on Jan 14, 2009 12:39 PM EST up reply actions
Back before the internet bubble burst
I was regularly involved in multi-million dollar deals with various high-roller types and the convention they used was M for million and MM for millions. Big whoop, eh?
Those Pilgrims ain't lookin' so proud now...
righto
giveml is correct. MM is million (a thousand thousands). It does fall under “big whoop” until someone puts one M instead of two in a contract and *crap*storm starts.
by Willie McGee's Twin on Jan 14, 2009 4:06 PM EST up reply actions
Why wouldn't you use one "M"?
It’s the first letter of the word. The second “M” is superfluous.
by Anonymous Communist on Jan 14, 2009 4:13 PM EST up reply actions
in this day and age
M is for Million
by Cards Fan in Chitown on Jan 14, 2009 4:52 PM EST up reply actions
M is the SI prefix for million
M = mega … MB=megabytes, MV= megavolts, etc.
"Dogs look up to us. Cats look down on us. Pigs treat us as equals." --Churchill
I always use
$1M, $2M, etc. For example, we signed Lohse to a 4 year $41M deal.
The Godfather himself has decided to grace us with his presence. This is his damn house. He sleeps 20 feet away.
by thegodfather on Jan 14, 2009 12:50 PM EST up reply actions
I get into this argument/discussion at work on a regular basis
Technically the Roman Numeral of “M” stands for 1,000 but almost everyone uses it for 1,000,000. The official roman numeral for 1,000,000 is a “M” with a line over it, but I’ve never seen anyone actually use it.
I typically go with common usage and use a M to represent 1,000,000 but every once in a while you run into an anal person wanting to show off how smart he is that throws a fit about it being used incorrectly.
okay..
Just for fun can we say it stands for one Miles which = $1million?
Sometimes I wonder,
"Why is that frisbee getting bigger?"
...and then it hits me!!
Not sure
Doesn’t one Miles now equal $2.4 or $2.6 depending on which year?
See my post above...
an average Miles is worth about $1M…the cubs are paying $5M for 2M’s…
Sometimes I wonder,
"Why is that frisbee getting bigger?"
...and then it hits me!!
M is Million
Logically, if we use K for a 1000 (Kilo), scientific notation dictates we use M (Mega) for 1 million..
But if some insist on using roman numerals, then it get’s confusing, and all bets are off.
I'd rather my sister be a prostitute than my brother a Cub fan.
And so, when someone says "megabucks"
they, literally, mean “millions of bucks”.
I have seen the MM “Roman numeral” designation for 1 million (although as pointed out above it is not the formal designation), but a single M seems to be the norm in most articles, mags, etc. that I read. In effect, since nearly every contract is in the millions, we could just not use anything. So, for example, Lohse signed a contract for 4/41 and the league minimum is .4 (or thereabouts, I’m too lazy to look it up right now). The fact that 41 stands for $41,000,000 is, more or less, obvious from the context.
by ArkansasTravs on Jan 14, 2009 10:41 PM EST up reply actions
I would be
happy taking a flyer on a Kris Benson type. Maybe Duncan can turn him into something. That’s a pretty long injury history though, isn’t it?
I really want to hold onto the draft pick. Sheets might be the exception if the Cards believe he can stay healthy, but I really look forward to the draft and to the opportunity to add high end talent to our system.
WOW the accuracy here..
…is just freaking scary.
bizarre question
what happens to draft compensation of a type a or b free agent agrees to a minor league contract, or can that even happen?
"No matter where you go, there you are" Buckeroo Bonzai Across the 8th Dimension
I don't think you can get around it that way
The only way to get around is if the originally team waves it or you wait for the draft.
Stat Whore
by FlimtotheFlam on Jan 14, 2009 1:01 PM EST up reply actions
Shannon Stewart signed with the Blue Jays on a Minor League contract
and screwed the A’s out of compensation.
"I'm on hold for now"- Bobby Crosby
by DyeLongJustice on Jan 14, 2009 1:19 PM EST up reply actions
GuitARF
I was ok ‘till I saw REO Speedwagon on the site. Brutal! I have some pretty hideous memories from the early 80’s that involve The Spudbucket. Although “Ridin’ the Storm Out” was kind of ok (but “heard it from a friend, who heard it from a friend…” is gag-me-with-a-smurf territory).
youneverknow
Seeing Tony LaRussa at a Metallica concert
would actually top seeing Lou Holtz at Preservation Hall in New Orleans last June…
"I just wish that the late Harry Caray were still around so I could hear him mispronounce 'Kosuke Fukudome' every fukun' night" -- Dennis Miller
I tell you what,
knowing that TLR went to an AC/DC concert definitely boosts MY respect level for the old man.
"Dogs look up to us. Cats look down on us. Pigs treat us as equals." --Churchill
jason jennings
is my nominee to replace the matt clement contract.
Now, it has been suggested that someone should round up all the projection systems, compare them to the actual performance of the players, and then try to determine how accurate these systems really are. Well, aside from the fact that I’m sure someone much smarter than me over at BP or the like has probably done so at some point
See THIS INSIDETHEBOOK.COM THREAD for link to a Nate Silver BP article which compares the following:
PECOTA, CHONE, ZiPS, Marcel, THT, ESPN Fantasy, Rotowire, and RotoTimes
The discussion in comments includes Tangotiger (Marcel), Nate Silver (PECOTA), and Rally (CHONE).
Projection Accuracy
What kind of makes things difficult is that the major projection systems are pretty close in accuracy, so you can generally make one system or another look a little better or worse if you seek it out. Not that these guys would do anything of the sort, but it’s just a natural human bias.
I try to stay out of these types of discussions. There’s just too much conflict of interest. While most of the forecasters know each other pretty well (Sean Smith, for instance, has been on BTF since it started in 2001), not all the people looking at projections will know who the hell any of those people are and whether or not they’re being self-serving or not.
So I take the lazy route with ZiPS. I do my own audits, but release nothing to the public, but I figure that with projections free and easily available going back to 2002, anyone’s free to do with the data as they wish (non-commercially, of course).
--
Dan Szymborski
dan@baseballprimer.com
by D.Szymborski on Jan 14, 2009 2:09 PM EST up reply actions
FWIW Nate put you right behind his own system :) Thanks for ZiPS and the response!
I understand if it presents a conflict for you, but I for one would love to see ZiPS at Fangraphs.
ZiPS at Fangraphs
When I talk about conflict of interest, I’m talking about the one in which I say X system is better than Y system when one of the systems is mine.
ZiPS will be up at Fangraphs. I’m just not finished yet. They were up there in 2007 and 2008 too. I love seeing ZiPS at Fangraphs because BTF isn’t really set up to be a stat site and I can get just about the presentation I want, but with David Appelman doing the work!
--
Dan Szymborski
dan@baseballprimer.com
by D.Szymborski on Jan 14, 2009 4:56 PM EST up reply actions
I wasn’t referencing the conflict of interest you declared – mine was a new statement about the understandable potential for conflict between sites, not projection systems. I’m elated to hear that there isn’t one, though, and will look for ZiPS at Fangraphs! I didn’t know that they were up in 2007 and 2008.
Conflict of Interest
I agree with Dan’s basic sentiment about conflict of interest. This is why I position Marcel as the minimum level of acceptance, and I HOPE that other system DO beat it. So, I have an almost anti-conflict of interest in that I try to position the study to give others the benefit of the doubt over Marcel.
That said, I am running this: http://tangotiger.net/forecast/
And Chone and Dan are active participants, as is Marcel and several others. PECOTA however is not. That means I’ll have to manually enter the data myself, and do the rankings myself. It’s a bummer of course, but I’m going to lay out all my work so that it is completely reproducible by the public at large.
Heh, I think if I had a projection system I might in dark times find myself frustrated with the regular, “But this is so simple!” sentiment :) I’m still pretty surprised that somebody hasn’t developed “Marcel+” by adding park neutralization or other tweaks.
VERY glad to hear about the Challenge! Thanks for the heads-up and for running it.
Dumpster Diving
I think virtually every club should engage in this great pasttime. Of course, it should be as part of a comprehensive, multi-pronged strategy and not the only source of players. I wonder who would make the all-dumpster team? I think Carpenter was a pretty good dive, especially if WJ had waited a little longer to offer that extension. Ludwick was obviously a nice dive and I think Suppan was pretty close to a dive wasn’t he? Then there was the Jeff Weaver 2006 phenomenon, Mike Matheney was pretty much a reclamation project, as well as Kent Bottenfield. I am sure there are others.
If anyone wants to make an all-Dumpster nomination please feel free.
Those Pilgrims ain't lookin' so proud now...
check fan post
june 14, 2008 for suggestions (scrap heap heroes)
"No matter where you go, there you are" Buckeroo Bonzai Across the 8th Dimension
"he gave up what is commonly referred to as a 'shit ton' of hits"
this cracked me up!
I also thought the Clement signing was a good idea… good thing he got Lohse as insurance at the last second. hopefully we’ll insure Carp similarly this season.
by Cards Fan in Chitown on Jan 14, 2009 2:07 PM EST reply actions
So here is the formula for projecting pitchers acording to VEB
4.4X era
around .500 records
5X walks
11X strikeouts
Wow...
didn’t notice how similar all three of those projections were. I don’t enjoy the pitcher projections nearly as much as the hitters…I think hitters are just easier to project.
Sometimes I wonder,
"Why is that frisbee getting bigger?"
...and then it hits me!!
Not really
Look at who we were projecting:
Injured Clement that could be his shutdown self or could implode, middle of the road was expected
Joel, showed good and bad the previous season, his StL experience was general decent, middle of the road expected
Lohse, has been fairly consistently middle of the road, thus that is what was expected
Reyes, high hopes from some, completely written off by others in the end we hoped he would recover to at least the middle
If these projections had been for Wainwright it would have been different. We also just had a lot of mid level pitchers
"People call me El Hombre," Pujols said. "But only Stan is the Man."
I don't know, the only one I would've expected to be middle of the road based on stats is Lohse (Hindsight!)
I think the rest was just not wanting to admit some guy is a bust, but not wanting to be a Kool-Aid drinker.
BTW, Shutdown or self implode does not = middle of the road. He’s not going to alternate between the 2 every other start. That’s kind of like going to a casino and saying, if I bet 50, i could win 100 or end up with nothing….so I’m going to guess I’ll end up with 50 because that’s the average of the 2 results.
obviously Tony must have thought Izturis was a good hitter
from Baron’s link
The idea of hitting a position hitter ninth only works if he is a good batter.
That part of the interview really cracked me up.
Also, that line sadly reads like some of my comments.
* sarcasm might be involved in this comment
This year
it will likely be Kennedy in the 9 spot…he should be an upgrade over Izturis.
with izturis's defense
him batting .263/.319 was just fine by me last year. If kennedy can do a little better hitting/obp combined with his defense at 2B, i’ll be fine with that too.
Cardwash - Cardinal, Washington fan (Washington???? Yeah, I know)
so is anyone going to go through the trouble
of determining how accurate the projections are? or maybe it’s already been done? I think I remember a post a while back where people suggested which was the most accurate. don’t remember which one was though.
by Cards Fan in Chitown on Jan 14, 2009 7:03 PM EST reply actions
oh yeah
I remember that article, thanks again!
by Cards Fan in Chitown on Jan 15, 2009 12:53 AM EST up reply actions
Sometimes I find the projections
a lot of fun. Other times, I just wanna see guys with gloves and bats out on a goddamn field, and don’t give a shit WHAT their numbers are. Today I’m in the latter mood.
by MdRedbirdFreak on Jan 14, 2009 10:09 PM EST reply actions
Speaking of Clement...
I thought I saw on the crawl across the bottom of the screen during the 2006 NLCS Game 7 replay the other night that someone had actually signed him for the coming year. It may have been a minor league contract, but I’m almost certain I saw that.

by 

















