The Miguel Tejada Signing & Different Roster Permutations

[I wrote this before it was announced that the Orioles had signed Miguel Tejada to a one year, $6 M contract – I was at FanFest when I found out and they didn’t have WiFi – but the post already contains my thoughts on the (potential) signing towards the end. I don’t like it, but if one assumes that the team was 100% assured to make the mistake of playing Atkins at first instead of third, then it’s OK. Tell me how you think Tejada will do for the O’s next year here.]

I keep hearing these days that the Orioles are looking for a third-baseman – possibly Joe Crede or Miguel Tejada – with the intention of having Garrett Atkins play first-base. I have to say that it doesn’t really matter all that much, as they’re just mainly buying time until Josh Bell and Brandon Snyder are ready, although it does indicate to me a less than fantastic thought process behind the team’s off-season actions. It’s also been reported that Nolan Reimold has been named the starting left-fielder.

For a team that really needs to squeeze every advantage out of the roster that they can, these things raise a couple questions. (I realize that the numbers I’m employing are all relatively small and close together, so the lack of precision can change the outcomes. Like I said; it doesn’t really matter much. It’s more the principle of the thing.)

So first up; why would you sign Akins specifically to play first-base?

There were several other (better) first-basemen on the market, so to jump on Atkins so quickly after he was released would have been not the best way to go about it in any case. Also, if Atkins is supposed to be a first-baseman then his contract looks even worse. At $4.5 M I could have lived with him at third. I wasn’t really happy about it – especailly given the way the market has turned out, with teams generally paying $3.5-4 M per Wins as opposed to $4.5 M – but fine. At first though, things (namely, Atkins) get worse.

For his career, Atkins has a -5.0 UZR/150 at third-base and a -6.3 UZR/150 at first. Jeff Zimmerman’s UZR/150 projections for 2010 has Atkins at -7 at third and -3 at first. Since the positional adjustment is +2.5 runs (at third) to -12.5 runs (at first) – a difference of 15 runs in a full season – Atkins would lose about 9-10 runs (~1 Win) off his value. That would make him less than a 1 Win Above Replacement player – closer to 0.5 WAR – which would be worth (in this market) probably only $1-2 M. His contract suddenly looks pretty horrible, and that’s without even thinking about the other better players the Orioles could have gotten instead.

If Andy MacPhail signed Garrett Atkins with the intent to use him as a first-baseman all along, then he made a mistake.

Let’s give him the benefit of the doubt though, and say that Andy’s plan was to pick up, say, Carlos Delgado for first, but it didn’t work out. Delgado apparently hasn’t recovered well from the hip injury, and it looks like he won’t be able to take the field all that often in 2010. So Plan A fell through. It happens; you can’t control everything; etc.

Plan B (sign a 3B and make Reimold the starting LF) is to use the roster suboptimally AND at greater cost? That should be Plan E (for extremely dumb*). Plans B through D should look something like:

* Clever!

D: Signing a guy like Hank Blalock or Russell Branyon for $2 M to give you something like 1 WAR of production at first. Then you have Atkins at third (~1.5 WAR) and come in at around 2.5 WAR from the two positions for ~$6.5 M. Decent.

C: Letting Michael Aubrey and Ty Wigginton (and whoever) split the time at first with probably about replacement level work. Then you have ~1.5 WAR for $4.5 M (basically just the original Atkins signing, but it frees up a roster spot and does less to block Bell/Snyder once they’re ready).

B: (My preference, if Scott isn’t traded – which is really my preference)

Letting Luke Scott and/or Nolan Reimold play first-base. I understand that it’s unclear how good either could be over there, but if you assume the alternative is the replacement-ish level work of Aubrey/Wiggy then you get the following scenarios for 1B, 3B, LF, and DH:

Plan C:

1B: Aubrey/Wiggy (0 WAR)
3B: Atkins (1.5 WAR)
LF: Reimold (say, 2 WAR with -3 run defense)
DH: Scott (say, 1.2 WAR).
Total: ~4.7 WAR (at +0 dollars relative to now)

(Pie fills in a little in left and in center, but in the other scenarios he has that playing time anyway)

Plan B:

1B: Scott – Say he can play league average first-base. Given that he’s an average-ish left-fielder you’d expect him to actually be a plus over there, but let’s just say +0. Then his value goes up to about 1.6 WAR.

3B: Atkins (1.5 WAR)

LF: Pie – Say 2.0 WAR with +7 defense, though we only count 1.5 of it here since he would have gotten the other 0.5 WAR in any case as the fourth outfielder.

DH: Reimold – Around 1.4-1.5 WAR. He loses about 8.5 runs due to the positional adjustment (from -7.5 to -17.5 going from LF to DH, but then multiplied by 85% for playing time) but gains 3 runs back since his glove won’t be a minus (it’ll be a zero).

Total: ~6 WAR (at +0 dollars relative to now)

You could do Pie in LF with Scott and Reimold switching just as well. It doesn’t really matter – you go with whoever is the better first-baseman of the latter two in the field, and I might be inclined to go with Scott there. He might have an easier time adjusting since he’s been in the big leagues longer; he played the position at least a little last year; it would build up his value for a trade (showing that he can still use his glove somewhat); and it might allow Reimold to heal from his Achilles injury easier by not having to run around on it so much. Obviously he (really, either guy) can fill in in left sometimes but Pie is easily the best left-fielder on the team (amongst guys that aren’t already in entrenched position in the outfield*). That’s something to take advantage of, given that Felix can hit a little too.

* Zimmerman’s UZR projections have Felix at +4 in LF and +2 in center, with Adam Jones at +1 in center. If Jones could put up at least a +3 in left – and I feel pretty confident saying he could – then it might actually make sense to put the him there and Pie in center where I think everyone can agree he is better, relatively speaking (that is, he’s a better CF than you would expect given his abilities in LF).

What you basically gain is not Reimold vs. Pie or Scott vs. Reimold or Scott vs. Pie, but Pie vs. Aubrey/Wiggy. And that is a deal I’d like to make; it’s basically free production!

Instead, the team sounds like it wants to go with Plan E:

1B: Atkins (0.5 WAR)
3B: Crede/Tejada (say, 1.7-2.0 WAR)
LF: Reimold (2 WAR)
DH: Scott (1.2 WAR)
Total: ~5.6 WAR (at plus $4-8 M, relative to now)

This improves the team over doing nothing and also not playing Pie more by only 0.9 Wins, at the cost of at least a few million bucks. In that case – if you’re already going to be making a mistake – it’s not terrible. I’d prefer Crede if you can guarantee* the three months for which he’ll be healthy will occur at the beginning of the year before Bell is ready. Otherwise, Tejada might be fine at $6-7 M.

* Through magic of some sort, perhaps.

On the other hand, this is still worse than letting Reimold, Scott, and Pie all start and at added cost as well.

The signing a cheap 1B one (Plan D) comes out as:

1B: Branyon/Blalock (1 WAR)
3B: Atkins (1.5 WAR)
LF: Reimold (2 WAR)
DH: Scott (1.2 WAR)
Total: ~5.7 WAR (at plus $2 M, relative to now)

That’s only better than signing another third-baseman by 0.3 WAR on the higher end, but it’s probably a couple million dollars cheaper at least. It’s also better than sticking with what they have but not playing Pie (which is the worst permutation), but worse than sticking with what they have and playing Pie (which is the best – and at this point, the cheapest – alternative).

The moral of the story is: Felix Pie has shown he can play a little, and he might be able to do even better than that. Let’s find out.


Just as a quick hit on something Chris at Baltimore Sports and Life said about this whole kerfuffle regarding signing Tejada to actually take Cesar Izturis’ place at short:

Tejada’s something like a -5 defender with an average-ish bat (maybe around .295/.330/.430). That makes him around a 2 WAR player.

Izzy can’t hit (around -20 runs offensively), but is more like a +10 defender. That makes him about a 1.5 WAR player in a full season, but only maybe a 1 WAR guy since it’s expected he’d miss some time.

If you assume that Cesar’s salary ($2.6 M) is already a sunk cost then you really shouldn’t pay Tejada more than $3-4 M or so for that upgrade from 1 to 2 wins at short (so you’re paying $6-7 M overall for the two wins). On the other hand, if Izzy could be dealt somewhere for something decent then giving Tejada that whole $6-7 M for one year wouldn’t be terrible. I don’t see who would have real interest in Izturis though, so while it would be an improvement to the team it’s probably not worth the cost.