I know I'm shooting up the charts with a bullet on the "Most Irritating Member of the Cafe" list, but I can't help myself, so here goes some more research findings.
While looking through some of the mock drafts here I realized that all veterans fall into 3 categories come draft day: Those who improved their value over the previous season, those who had their value drop over the previous season, and those who maintained their value from the previous season. Nothing revolutionary here, I know, but I got to thinking, which of these groups are the safest to pick come draft day?
So, what I did was take a random sampling of those players who played on '01 and '02 and placed them in three categories: Those who had a better year in '02 than '01 (an increase of production of at least $5), those who had a worse year in '02 than '01 (a drop of production of at least $5) and those who stayed about the same. Then I compared these values to how they did in '03 to try to find which group of player would be safest to draft. This is what I found:
1) Those who had better years in '02 than '01 (23 in my sampling), only 3 found further improvement in '03 (Rentaria, Holliday, and Prior), 5 had equal production, and 15 had worse seasons than they did in '02. This is a pretty risky group to pick from, with only a 30% chance of drafting a player who will equal or better their prior season's numbers and 70% who'll do worse. What makes it trickier is that, although young players seem to be a better bet for continued improvement, it is still a poor gamble to take. Of those players in this cat who were under 30 at the beginning of the '03 season, only 3 improved and 2 stayed the same. Some players who will fall into this category come '04 that you may want to be leary of are Pujols, Helton, Wilson, Delgado, Prior, Wells, and Sheffield.
2)Those who did worse in '02 than '01. This is the group we all should be focusing on in my opinion. Of the 19 playes in this group, 10, saw their numbers improve by at least a value of $5 or more in '03. That's better than 50%. And only 3 of the 19 continued their slide (R. Alomar, Erstad, and Griffey). 6 repeated their performances from '02. Look at some of the players we could have plucked out of the '03 draft who fall into this group: I-Rod, Helton, Mussina, Vazquez, B. Boone, Sheffield, Kendall, Hudson, and Pierre. That's a championship team right there. This is the group that wins you championships in my opinion. Some players who will fall in this group in '04 are Guerrero, A-Rod, Bonds, Ordonez, Soriano, Jason Giambi, R. Johnson, Berkman and SHAWN GREEN (!!!!!!!!!!!!!!!). Keep in mind though that players such as Bonds and Johnson may be reaching the age were improvement may not be possible, but I wouldn't put it past them.
3) Ths group contains players who's '01 and '02 seasons were fairly equal. In '03, only 1 showed improvement (Pujols), 6 maintaned that level, and 14 had their '03 seasons tail off. Of those 14, though, all had their seasons marred by injury, so I don't know what you can make of it other than it's hard to to have three injury-free seasons in a row.
It may not be totally accurate, but I believe it is telling none the less. You fellas can do with this what you will, but I think it's at least an interesting topic.
Oops! I discovered an error. Rentaria found himself in the wrong grouping. He should have been in group 2 and not group one. He regressed in '02 compared to '01. This further substantiates my findings, though.
I think it's important to note that where these people are drafted greatly matters as well. You're working with such a small sample size that your numbers are somewhat meaningless. But the fact that the "bad year" group seemingly improve and you can draft them late is good to note.
A word of caution: A lot of your analyses posted here are based on a very small sample. Beware...
Erboes wrote:I know I'm shooting up the charts with a bullet on the "Most Irritating Member of the Cafe" list, but I can't help myself, so here goes some more research findings.
Don't worry most of the best members are at least a little irritating, myself included.
Erboes wrote:It may not be totally accurate, but I believe it is telling none the less. You fellas can do with this what you will, but I think it's at least an interesting topic.
It is interesting but you need to know to what degree a player with improve or decline. Renteria for example will be hard pressed to improve or match last year's numbers. Jeter if healthy will improve. Who is going to have a better year?
If pollsters can use 900 of a possible 280 million people as a scientific sampling can't I use 25% of a possible 256 in a standard 12 team, 23 player per team league? Your warning is true, of course. I think if I had the time or inclination going back two more years would make it more accurate, but I think the trend would stay the same -- the best values in any draft or auction are in the bounce back candidates.
Rentaria, for '04, would fall into the first category, which is players who improved in '03 compared to '02. So, if you believe my findings, he has a better than 50% chance of losing at least $5 of production for "04. If that happens, the best he can be is a $30 player. By my sampling, Jeter would now be a bounce back candidate and would have about a 50% chance of increasing his production in '04 by at least $5. If that happens, he would at least be a $24 player for next season. Judging by their career numbers, I think it's a very good bet that Jeter will put up better numbers than Rentaria next season and if I had my choice I'd take Jeter. Of course, Rentaria could be one of the few who improve over back to back seasons and Jeter could be one of those in decline, but like they saw when playing blackjack, you got to play the odds.
Erboes wrote:If pollsters can use 900 of a possible 280 million people as a scientific sampling can't I use 25% of a possible 256 in a standard 12 team, 23 player per team league?
This is misleading. It's been several years since I took statistics, but I think a poll of about 1,000 people gives you a sampling error of +/- 4-5 percent. If you increase your sample size, your margin of error only decreases minimally.
Your 25 percent number is inaccurate because you're using three years as a baseline--what the improvement/decrease in production over two years means for a third year. What about all the other years?
Erboes wrote:the best values in any draft or auction are in the bounce back candidates.
Absolutely. Just make sure you don't pay full price for them or jump the gun and draft them too early.
Yes doctor, I am sick. Sick of those who are spineless. Sick of those who feel self-entitled. Sick of those who are hypocrites. Yes doctor, an army is forming. Yes doctor, there will be a war. Yes doctor, there will be blood.....
I stated that to make it more accurate I'd have to go back a few more years, which I ain't going to do. This was enough to substantiate my belief that the best values out there are those players who are coming off years. There seems to be two kinds of drafters out there, those who think that a players previous years numbers will equal those in the next and those who don't. I'm one of those who don't. Green has been a second round type player for a few years now and one bad year makes him a sixth rounder? I don't buy into that thinking.
And what I am trying to do makes a players previous seasons in that 3 year cycle irrelevant. I'm just trying to show how players tend to improve or not depending on that three season sample and which players may or may not be good values because of it.
I am trying to substantiate my thinking with facts, that is all. Whether you believe them or not is none of my concern.
Erboes wrote:I am trying to substantiate my thinking with facts, that is all.
Not to jump on you, but this is a tough place to start. If you want to substantiate what you believe with numbers, you'll always suceed. You really need to start from an unbiased perspective to come up with an unbiased answer.
Nonetheless, I agree that the best value is in "bounce-back" players, mostly because you can get them late and the odds are that they will return to form. There are plenty of players that had huge years that were flukes. But then there are some that had break-out years and are here to stay. Deciding what category a player falls into is more complicated than stats. It often comes down to a makeup of a player, his team, etc.
Transmogrifier wrote:A word of caution: A lot of your analyses posted here are based on a very small sample. Beware...
Agreed!! I love looking for statistical trends too, but with small sample sizes and lots of parameters it's easy to read too much into trends you might find.
For example, what do you mean by 'random sampling' of the players you chose? Did you put all the names of players in the three categories in a hat and draw out 20? Or did you pick 20 names you were familiar with? For players with a decline in value, did you exclude those who were injured? I'm not asking to nitpick, but because depending on how you picked the names you may have introduced some kind of bias.
I do think you are correct that it is worth taking a look at players that underperformed what was expected of them. We need to look beyond the numbers to WHY they underperformed, and that is certainly a good place to look for draft day bargains. There are lots of great bargains to be had in many leagues, and it's good to keep them in mind on draft day - and perhaps even to talk them down to discourage others from drafting them.
From your group 2:
Guerrero: Probably not much of a drop in draft day value.
Giambi: slight drop to late 1st, early 2nd due to knee injury. Probably warranted.
Ordonez: Might see slight drop in draft position. He will be 30 entering next season, and may be tapering off due to age. His 2002 numbers are much better than any other year. His 2003 numbers are much closer to his typical production (though his RBIs were down a bit, he actually posted his 2nd best career OPS). Expect him to post similar numbers in 2004.
Soriano: Some people might be scared away due to his post-season performance and KC trade rumors. Probably not enough for him to slip to a bargain spot on draft day. He'll probably still go top 5 overall.
RJ: As we've seen in the mock drafts, he could fall a lot. Definitely someone to keep an eye out for on draft day, but no longer a 1st rounder.
Berkman: Took a step backward, but was playing through little injuries. He could be a draft day steal. Get him in the 3rd round if he's still there.
Green: Tough to say. How can we trust that he's healthy? Last year he quietly played through a major shoulder injury (good move, management!). Worth a risk, but only after you've got a core (i.e., I wouldn't bet the farm on him).
I think you have stumbled upon one really interesting question: How many players can maintain their value over a three-year period? And how many of those players are top-tier players? I haven't looked at the numbers, but my guess is: not many. That's why we pay a premium for consistency in fantasy baseball, and that's why someone like RJ is no longer a first-rounder.
"The game has a cleanness. If you do a good job, the numbers say so. You don't have to ask anyone or play politics. You don't have to wait for the reviews." - Sandy Koufax