Reese was signed to be a starter in 2005, because the team didn't feel that Jose Lopez would be the answer at shortstop. This much we know. There are questions about this, of course - is it because they don't want to rush Lopez, or because they want him at another position? Do they think that Reese will be a better contributor than Lopez this year, or do they just want to keep the kid from being overmatched? Nevertheless, we won't get a definitive answer to them any time soon, as the team's handling of Ramon Santiago so far in ST suggests that they're willing to go to considerable lengths to keep Lopez in the minors.
The argument about service time and rushing Lopez is much the same here as it is with Felix Hernandez. The effects of eating up another year of Lopez's clock won't be felt for a while, and nobody knows what condition the team will be in four or five years down the road, so it becomes an argument based in the abstract and hypothetical that is quite difficult to make.
There is one difference between the Lopez and Hernandez cases, though: where we can be pretty certain that Hernandez would be an improvement over Sele/Franklin out of the rotation this year, the difference between Lopez and Reese isn't so clear. Reese is arguably the best defensive infielder in the league, where Lopez doesn't really belong at short, and the offensive gap between the two may not be large enough to make up for the defensive difference, especially if Lopez repeats his 2004 performance.
The only thing we can do in this situation, lacking 2005 statistics, is to project performance and see what we get.
According to PECOTA, Pokey Reese is a very common player, in that there exists a large group of past and present players with similar physical and performance profiles. With that in mind, Reese's PECOTA projection may be more considered more accurate than that of, say, Randy Johnson, to whom there have been few (if any) comparables.
The system projects Reese to put up a .222 EqA in 190 plate appearances (it didn't expect him to be an everyday player, which - given his medical track record - he probably won't be).
On the other hand, PECOTA predicts Jose Lopez to put up a .260 EqA in 517 plate appearances (it obviously expected him to be the starter for most of the season).
For the sake of keeping things simple, let's say that - as the starting shortstop - Reese takes those 517 plate appearances. A .222 EqA over that sample would yield about 43 runs created with the bat.
For Lopez, a .260 EqA over 517 plate appearances would yield about 61 runs created offensively.
So, given equal playing time (about 136 games), PECOTA projects Lopez to be worth somewhere around +18 runs over Reese at the plate.
Now, for Reese to balance out that difference, he has to be worth at least +18 runs over Lopez in the field. What we get from 2000-2003 UZR data (that's all I've got) is that Reese is worth 8 runs in the field as a shortstop. However, this is based entirely on a half season at the position in 2001; a sample more recent and four times as large has Reese at +22 runs defensively as a second baseman, which isn't that different of a position. The truth probably lies somewhere in the middle - splitting the difference puts Reese at +15 with the glove as a shortstop per 162 games. Narrowing this down to our 136-games estimate for 2005, we get that Reese should be worth approximately 12-13 runs with the glove in 2005.
It's worth noting that, although the UZR data doesn't extent to 2004, David Pinto's Probabilistic Model of Range numbers showed that Reese had the best range at shortstop in the Majors last year.
As for Lopez's defense, it becomes something of an educated guessing game, since 2004 UZR numbers haven't been published and he didn't accumulate enough playing time to show up in Pinto's system. With that said, it's probably safe to assume that Lopez would be a below-average defensive shortstop in 2005, based on the scouting reports and numbers we do have - BP's metrics show that he was an incredible 10 runs below average in the 57 games he played at short last year. Thus, it's not inconceivable that, over 136 games, Lopez would be at least five or six runs below average with the glove, enough to balance out the offensive difference between himself and Reese.
Of course, this doesn't say it all - there aren't too many players comparable to Lopez, which sheds some doubt on his PECOTA projection. As a 21 year old power-hitting infielder, it's quite possible that Lopez could make "the leap" this year and put up an OPS around .800. He has offensive potential that you just don't get with Pokey Reese. That said, it's also possible that he could repeat his dismal .232/.263/.367 line from the second half of 2004, which would make him a worse hitter than Reese without the awesome glove. Where you have a pretty good idea what you'll get from Reese at the plate, Lopez could fall anywhere between two extremes.
That said, Lopez's minor league numbers over the past three years have consistently translated to a decent-average, low-OBP, respectable-power Major League line, hovering around .265/.305/.420. Assuming that he continues along with his development pattern, Lopez's .268/.304/.421 PECOTA projection seems pretty solid. If he achieves that level of performance, then the difference between Lopez and Reese is very small, if not completely negligible.
There's been a lot of resistance to the idea of starting Reese and demoting Lopez because offensive productivity is obvious - Reese sucks at the plate, and this is easy for anyone to see. What he adds to the team isn't nearly as clear, because defense is tough, if not impossible, to assess with the naked eye (see: Derek Jeter). This, along with the fact that defensive metrics haven't come as far as offensive statistics, has fueled the pro-Lopez argument ever since the day Reese was signed.
The calculations I've run through here aren't perfect, and there are certainly little bits and pieces that can be picked apart by anyone so inclined. All I mean to do is introduce the pro-Reese side of the debate, because there have been few people willing to do such a thing so far this winter. It's up to you to decide whether or not you're convinced by either argument, and a lot of this hinges on how much faith you have in defensive statistics. Whichever way you lean, though, understand that Reese will make a positive contribution to the team as the best defensive player on the roster. He may not be Jose Lopez with the bat, but nobody is Pokey Reese with the glove.