Evaluating ST Performances

[endgame]

Concluded ST a day or two ago, and uncharacteristically in my tenure with OOTP I didn't watch every game, but instead kept an eye on news and the Free Agent market, looking for potential minor league talent. It surprised me a little to find it was over and now, task at hand, to the review of players on the roster, their performances, and potential to be part of my active roster.

It's a pretty competitive 40-man, with a lot of players within inches of each other in 'star' ratings and potential, so I'm left to look at my scout's view of their specific ratings and their spring training performances. Granted, it's a phase designed to shake off the rust, etc... but how much weight do most of you actually put on the ST results? For example, if you had a highly rated MR who actually tanked during ST, would you move forward with him, and just monitor his first 10-20 game appearances, or consider filling a slot with a not-so-highly-rated who really showed his stuff during ST?

Just curious. In the past, I've often just cut rosters almost by star ratings alone, but now that I'm in this position, I'm don't know how far I want to take my evaluations toward a stats-only consideration, given it's ST play.

What cuts it for you? Thanks in advance.

[eriqjaffe]

I usually have a pretty good idea of who is going to make the 25-man roster, and those "bubble" decisions are often based on who has minor league options left. But on those occasions where that's not an issue, I wind up giving a pretty good amount of weight to the ST stats, especially if the players appear to be very similar.

[RchW]

Originally Posted by eriqjaffe

I usually have a pretty good idea of who is going to make the 25-man roster, and those "bubble" decisions are often based on who has minor league options left. But on those occasions where that's not an issue, I wind up giving a pretty good amount of weight to the ST stats, especially if the players appear to be very similar.

Very similar to my approach.

[Fyrestorm3]

I put VERY little stock in ST stats. The only time I pay much attention to them is when I have two minor leaguers fighting for a roster spot. Otherwise, even though I still go through the exercise of evaluating and grading each player's performance, I can't bring myself to demote or promote someone based on ST stats alone.

Case in point: My ace pitcher, who's won 15+ games in each of the last 3 seasons, and took the Cy Young once, has consistently put up horrendous numbers in ST. His entire career, he's never looked like an ace in March. But come April, he's always back to form.

[soxfan34]

Originally Posted by Fyrestorm3

I put VERY little stock in ST stats. The only time I pay much attention to them is when I have two minor leaguers fighting for a roster spot. Otherwise, even though I still go through the exercise of evaluating and grading each player's performance, I can't bring myself to demote or promote someone based on ST stats alone.

Case in point: My ace pitcher, who's won 15+ games in each of the last 3 seasons, and took the Cy Young once, has consistently put up horrendous numbers in ST. His entire career, he's never looked like an ace in March. But come April, he's always back to form.

Agree

[D-BacksJosh]

Originally Posted by Fyrestorm3

I put VERY little stock in ST stats. The only time I pay much attention to them is when I have two minor leaguers fighting for a roster spot. Otherwise, even though I still go through the exercise of evaluating and grading each player's performance, I can't bring myself to demote or promote someone based on ST stats alone.

Case in point: My ace pitcher, who's won 15+ games in each of the last 3 seasons, and took the Cy Young once, has consistently put up horrendous numbers in ST. His entire career, he's never looked like an ace in March. But come April, he's always back to form.

I second that. ST to me in OOTP is just 'shaking off the rust' and I only use it for evaluation purposes as a last resort. The results in ST compared to regular season are too inconsistent. Much like real life!

[MizzouRah]

Originally Posted by eriqjaffe

I usually have a pretty good idea of who is going to make the 25-man roster, and those "bubble" decisions are often based on who has minor league options left. But on those occasions where that's not an issue, I wind up giving a pretty good amount of weight to the ST stats, especially if the players appear to be very similar.

I do the same.. mostly it's only evaluating MR's.

[Sweed]

Originally Posted by eriqjaffe

I usually have a pretty good idea of who is going to make the 25-man roster, and those "bubble" decisions are often based on who has minor league options left. But on those occasions where that's not an issue, I wind up giving a pretty good amount of weight to the ST stats, especially if the players appear to be very similar.

Pretty much the same for me. I have my "already going to be on the opening day rooster no matter what" players which doesn't usually leave many openings for guys to compete for. My everyday eight players are almost always a given too.

Like eriqjaffe, and a lot of real life decisions, it comes down to options if all other things are relatively equal. A talented player, determined by scouting and minor league history, that has a bad spring but is out of options will get the chance before a like player that still has options even if the second player had a better spring. If he doesn't show me something at the MLB level it's time to trade and bring up the second guy.

If I have two guys with comparable talent, career minor league stats, and options remaining then I will let one of them earn the spot with their spring training performance.

I do play out all spring training games. Like real life it really helps one to get a feel for the organization and how close that youngster you've been waiting on for the last three years is to finally making the club. It is one of the things I really enjoy about OOTP. I couldn't do it any other way.

[Nunyer]

Spring Training are the only games of the year that I consistently just simulate through. Spots 26 thru 40 on most rosters are so awful that I find the stats are just too skewed across the board to offer much. At best, the results are used as a tie breaker to decide between my last bullpen arm / final bench seat.

[hluraven]

I have never once looked at ST stats, results or anything else about them. I just go through it hoping no-one gets hurt. Too small a sample size and not "real" competitive fixtures, so any stats are meaningless to me.

[braunfullyaccused]

Originally Posted by hluraven

I have never once looked at ST stats, results or anything else about them. I just go through it hoping no-one gets hurt. Too small a sample size and not "real" competitive fixtures, so any stats are meaningless to me.

Although there's no right or wrong way to play this great game, I feel the statement, "too small a sample size" may certainly be a matter of context." I completely understand how standard deviation works, but for those of us that play every game versus the simulation guys, ST is a significant period.

Very similar to real life, given the time investment taken in the more real time approach to playing OOTP, I will most likely value the results of smaller periods higher then the sim guys who will see 25+ seasons over a few months or a year. Thus, I view ST as important and the results hold more weight in my decision making.

Much in the way watching the same player continually strike out with the bases loaded while playing games usually tends to impact my decision making beyond a strict statistical standpoint. I've found animosity goes a long way.

[RchW]

Originally Posted by braunfullyaccused

Although there's no right or wrong way to play this great game, I feel the statement, "too small a sample size" may certainly be a matter of context." I completely understand how standard deviation works, but for those of us that play every game versus the simulation guys, ST is a significant period.

I think you may be assuming too much about what simmers do. I sim games instead of playing them out but otherwise spend copious time analyzing each game or series of games. I use ST to help firm up decisions already in progress ie a trade, a position change or who the backup catcher will be. I look at other teams. However the most critical decision is whether a new player makes the team over an incumbent (starter or bench player; relief pitcher or starter). As erikjaffe and Sweed said options can affect these decisions which may prompt other actions. ST is an excellent time to trade players which I often do in the last week.

My goal is to have no players worth claiming on waivers at the end of ST. Not always possible.

[The Game]

I sim ST most of the time. I think the AI does a better job of getting certain players playing time then I might. I do pay attention to stats but only for "bubble" guys. For my current league i used a 6 man rotation in ST strict order. I had my 5 man rotation set but i wanted to get a few guys some ST starts in the 6th spot and see if they would have enough to be long relief during regular season. I had 19 pitchers in camp. I use a 12 man staff for this team, for now. i kept one guy who had 21IP 34 K 4 BB. So far he is borderline going down after 5 IP 9 BB 2 K. On offense I had a battle for OD 2B. I used 6 2B in ST all are young players or former ML fringe players. The 2 I kept had the best springs and although their OBP is decent 15 games into the regular season neither are hitting. My #1 SP, a clone of Greg Maddux had a terrible ST .345 OPP AVG. 7.36 ERA. He has given up 17 hits in 27 IP so far this regular season with a 1.86 ERA. So I dont put much stock into a veteran player having a bad ST.

[The Wolf]

I sim through ST because ST is there to remove player rust and the AI does a fine job of doing that and getting everyone playing time. I even do with no pitching roles or lineups.

After ST the wins and losses and stats actually count, so then it's time to get real. Not before.

[Le Grande Orange]

Originally Posted by braunfullyaccused

Although there's no right or wrong way to play this great game, I feel the statement, "too small a sample size" may certainly be a matter of context."

In regards sample size, I would point out that, as compared to real life, OOTP's spring training schedule is about 25% too short.

If you count split squad games and games against minor league and college teams, an MLB club in real life is typically set to play between 30-35 games in spring training. In OOTP it's only 24 games. Indeed, the median number of spring training games scheduled several years back was 32. Recently, however, that figure has gone up to 33, then 34, and in this year's spring training it was 35 games if I recall correctly.

Now, maybe an extra 8-11 games' worth of spring training stats in OOTP wouldn't make much difference. But it would be a larger statistical sample size.

[RchW]

Typical real life ST rosters are much bigger. I'd submit that the necessary players get more than enough AB or IP in OOTP ST.

[Le Grande Orange]

Originally Posted by RchW

Typical real life ST rosters are much bigger. I'd submit that the necessary players get more than enough AB or IP in OOTP ST.

They start off larger but the cut downs come quickly. Even without split-squad and games against minor league and college teams the number of games scheduled is still on average around 30-32 meaning OOTP is about 20% short.

[braunfullyaccused]

My comment was in no means meant to disparage the gents who sim or even take a hybrid approach.

My point was that by playing every game and the associated time investment required – doing so begets a certain element of player specific feelings that I feel may be lost through simulation play.

I believe Spring Training, regardless of length parallels that concept, but even more on a microscopic level. Similar to how seeing prospects fail or succeed with two outs and RISP during Sept. call-ups, the lines between statistical performance, scouting ratings and in-game performance begin to blur together as they relate to decision making - something that isn't as transparent in simulation play.

Again, there is no right or wrong way to play. I'm just relating the concept of 'time' and the influences of the aforementioned 'stock'.

[cavebutter]

I'm a simmer (except for pennant race and playoff games), but pore over daily box scores and stats. I look at ST box scores just as closely and, like others have said, use ST stats to help make bubble decisions. I use each off-day in ST to cut about 1/3 of the active roster. This gives players with a more realistic shot at the opening day roster more playing time- and more stats. Option years come into my decision-making, but it's not the main factor. I like trading during ST and I'm just as likely to trade a guy out of options as I am to waive him.

[beorn]

As best as I've ever seen, the #1 error made by OOTP managers is making changes based on small sample sizes.

If I have a sense of a player's quality over a whole season, I'm not going to change it because of a few ABs or IPs in spring training.