Name, Yr. drafted Draft Rd SF Years SF Games Played OF Starts
Nate Schierholtz (2003) 2 (63) 2007-12 503 258
Fred Lewis (2002) 2 (66) 2006-09 326 214
Todd Linden (2001) 1-s (41) 2003-07 270 79
John Bowker (2004) 3 (100) 2008-10 183 46
Dan Ortmeier (2002) 3 (97) 2005-08 124 33
Jarrett Parker (2010) 2 (74) 2015-current 84 39
Mac Williamson (2012) 3 (115) 2015-current 64 35
Roger Kieschnick (2008) 3 (82) 2013 38 21
Gary Brown (2010) 1 (24) 2014 7 1
Wendell Fairley (2007) 1 (29) — 0 0
Eddie Martinez-Esteve (2004) 2 (70) — 0 0
Dylan Davis (2014) 3 (87) Still in system
Bryan Reynolds (2016) 2 (59) Still in system
Heath Quinn (2016) 3 (95) Still in system
ogc thoughts
First off, I would note that the level of conversation regarding the Giants, i.e. Sabean, and the draft has been narrowing over the years. First, it was that he didn't know how to draft. I was one of those, and that sparked me to research it (see the study I produced on that here) and discover that the Giants were no more worse or better than teams that were considered good at drafting, and then I researched the draft itself and found that it is extremely hard to find a good baseball player even in the first round. Since then, I have been a fervent Sabean supporter, garnering the enmity of other Giants fans who fervently disagreed. They particularly didn't like it much when I would jump in and be thankful that Sabean got another two year extension while over 99.9999% of the comments in the thread where aghast that he was signed again.
Once he started getting success (which I had found in my study to be more likely once we started losing), then the Naysayers had to back off of their statements. Once Cain and especially Lincecum came to the fore, they couldn't say that he can't draft and develop starting pitchers. Romo killed any mention of relief pitchers. They moved on to position players, and Posey killed that one. Then Belt and Crawford killed the thought that the Giants couldn't develop middle infielders or corner infielders. So that left outfielders to hang their hate on (and lately, the conversation has turned back again that Giants can't develop starting pitchers, since Bumgarner was the last one in 2010).
Research Says: It's a Work in Progress Because Context Matters
I understand why people have problems believing that Sabean has actually been good with the draft. The problem is that the odds are so bad even in the first round. Look at the study I did on the draft that is linked above. Here are the key results I want to point out here (where a good player was defined as a player providing roughly 6 years worth of good production, either .295+ BA or 3.30- ERA; best data available online at that time; I've been hoping to update using WAR):
Star/Good: The data grouped into these sets: 1-5, 6-20, 21-30, 31-90, 91-100. The percentage of picks in each set, respectively, are: 43.1%, 18.5%, 10.8%, 4.0%, and 1.5%.So beyond the first five picks of the draft, the odds drop precipitously on finding at least a good player, even worse than the Mendoza Line, and once you get to the draft position where playoff contenders usually get, it just gets exponentially worse from there. By around the end of the third round (depends on how many supplemental picks are awarded), picks 91-100 become a good player or better only 1.5% of the time, per my study's results.
So while the Giants picks have gone sour, the odds are that they will. The highest pick spent on an OF was 24 for Gary Brown, and even there the odds were around 11%. You would have to pick around 9 Gary Browns to get to the point where you wonder if the drafter is any good, though as we'll see later, even then, maybe not.
Meanwhile, the Giants have spent their best draft bullets on these players: Posey (5th pick), Wheeler (6th), Lincecum (10th), Bumgarner (10th). As you can see, when you have better odds of finding a good player, your results look pretty good (and I would note that this string of picks was probably historically good in terms of impact). And when teams on average don't do so well in finding good players, then your chances of bucking the odds are pretty low as well.
Understanding the Implications of Such Low Draft Probabilities
One way of judging how good a team or GM has been with their draft picks would be to add up the percentages in the study's results above. Thus, we would count 4% for the Schierholtz pick, 1.5% for Bowker, 11% for Fairley, and so on. Doing this, the total for the players drafted above equal 60.0%.
What does this mean? It means that if you were able to replicate this an infinity number of times, with a similar set of players drafted each time, 60% of the time the drafted players will yield a good player. Thus, even though 14 players were drafted, the odds were not much better than a coin flip, the drafter would need to draft another similar set of players another 9-10 times in order to get to 100% No wonder so many of them has gone sour!
Of course, I should note here that reach 100% does not guarantee you that you will find a good player after that many picks. It is like rolling dice and counting up each time a certain number comes up. If you do it a significant number of times, you will find that a 7 will come up 16.667% of the time (or 1 out of 6) but if you try 36 throws of the dice, while 7 be rolled 6 times often, there will be many times where your "luck" is bad and randomness wins the day and you might end up with nothing.
With the draft, a similar concept holds. Sabean and gang could draft another 10 outfielders in the same range of the draft and it would still be within the bounds of randomness that he won't find a good starting outfielder and it don't mean that he's not good at drafting. For example, let's say Sabean ends up getting the 21st pick for the next 9 years/drafts and use 11% as the proportion of picks that become a good player. In 9 years, that works out to roughly 100%. 39% of the time, he would find one good OF. And 26% of the time he will end up with more than one good OF. However, 35% of the time he would end up with not even one good OF, and yet he could be performing at average. In other words, over a third of the time, he could still find no good OF and could still be good, just unlucky.
Haters Gonna Hate
So how people come across when they look at the draft tells me a lot about their preconceived thoughts about the GM. If they just love the GM, they will just say he's great, look at the rings and bow. If they don't care for the GM, they will point out any negative that they can find about his drafting.
Hence the sequence over the years - "He's horrible drafting..." "He can't draft position players..." "He can't draft middle infielders..." "He can't draft outfielders..." and now they are cycling back to "He can't draft pitchers...". Or they would point out all the bad trades, ignoring the fact that it didn't cost us anything much in terms of prospects to pick up the guy, and more importantly, not properly weighting the impact of trades (yes, GM's do make mistakes, but if he picks up great guys regularly, that should get extra weight).
And Generalists Gonna Be Too General
And sometimes, they just don't know enough to say, but say it anyway because the demands of the job require it. That is why my stance has been that the best analysis you are going to find about any particular team is going to be done by people who are focused on that team, because they care about the minutia about the team, whereas the generalists just don't have that breadth of knowledge about the team.
And the author of the article that started this blog post, he's both a generalist (worse, he followed the other Bay Area team for many years) and a hater (he's always been rough on the Giants over the years). He's a smart guy, but out of his depths when throw in by his editor to write on the Giants. And if he's not a hater, he's worse: a writer who don't know what he's writing about.
GM Analysis
This is why I've been saying for over ten years now that one can't look at draft results and say whether the GM is good or not. Especially not within the lifetime of most GM's, because they often don't last very long at the job. Especially since, once you get past the first 5-10 picks of the draft, it takes the rest of the players 4-6 years just to make the majors, let alone start to produce and demonstrate how good they are. The odds against and the length of time just to make the majors and be good just makes it basically impossible.
This is also why I've been saying for over ten years that no GM should hope to build a team from only home grown players (and thus no fan should denigrate a GM for not building a team with only homegrown players; in this case, the fan's hate is showing, no team has ever built entirely with homegrown and won). It is hard enough just to find good players for any particular position, based on the poor odds of finding good players, and thus almost impossible to try to game the randomness and find and fill a specific position. One should celebrate the wins (when there are enough of them) and not get so hung up on filling any one position or grouped positions.
This is why I wrote in my business plan series about how the best way to build up a team from scratch is to focus most of your top picks and the majority of your picks overall on pitchers, because the cream will rise to the top, as there are 13 positions related to pitchers, and with each good pitcher you find, you can fill out the rotation, as well as the bullpen. And while you keep the cream, the middle tier is also great for trading for the pieces you need.
Which is how you can build out a team beyond the home grown good players. And trading is the third leg of the system that has to be done well. And the skill involved with that is also shown by picking up minor league free agents who can contribute to the team at a cheap salary.
That's why I think the best way to evaluate a GM is looking for some key features of his team. First of all, he has a lot of homegrown good starters (above 2 WAR), more than other playoff competitive teams. That, to me, is a sign of a good drafter, that he's able to find a lot of good starters through the draft, relative to the league. Secondly, he's capable enough to fill in the rest of the roster, within the limitations of the budget and trading/free agency, such that the team is good enough to make the playoffs for many years, while they are together. And to win it all, while doing all the above.
Giants and OF
So the outfield is a work in progress. While it seems like we have drafted a lot of them with high draft picks, the fact is that it exponentially hard to find and draft players via the draft, and the Giants have just not devoted many higher draft picks on the OF. Pitchers yes, catchers too, and shortstops as well. These are often players who can move into multiple spots in any 25-man roster.
We have a number of interesting OF prospects, and calling the results sour just reflects the record to now. A while back, they could (and did) complain about the draft, only for the Giants to develop someone to make the complaint moot. As the Mercury's companion article noted, their drought might be over, with Parker and Williamson both battling it out for the LF starting position, and both doing so well so far. And as well, Davis, Reynolds, and Quinn are interesting prospects with potential.
I'm not too worried about the OF. We have a number of interesting prospects there but it is just so tough to make the majors. More importantly to me, we have at least a couple of players who look like they will be good starters in the majors in Tyler Beede and Christian Arroyo.
Too busy to research it, but it seems to me that there were more a few other outfielders in our system that played a few cames here and there that were drafted in our system.
ReplyDeleteYou are right, and that is because the author focused only on the top draft picks, I think the first 3 rounds, for this article.
DeleteHe did not state, but the thinking probably was along the lines of "beyond the third round, any success is related to luck, not skill", for as I noted, I think his intention was to make Sabean look bad. Or maybe it was more benign and he didn't want this huge list taking up space. Or a combo of both.
The Giants haven't spent many high draft picks on the OF, as you say, ogc. I am guessing that the reason for this is that they want to beat the odds that you've presented, therefore to look to the positions where shrewd scouting counts most; and for position players, that means looking to players who excel in the field rather than at the plate. One's arm, one's range, one's agility, one's speed, and one's hands are tested by amateur ball at a level that often approximates the professional level. Smart scouting can assess them well. Hitting against wild, undeveloped arms with narrow pitch repertories would seem to me much harder to assess. It would follow, I would think, that positions where the bat is more crucial--corner outfield and first base--would be harder to predict success in, whereas those where one can see skills tested by hard-hit balls and tricky fielding chances would be more projectible from high-school or college ball into pro ball. It's not an accident that the Giants have great defensive teams of home-grown players, but the result of smart drafting based on smart scouting, so as to overcome the draft odds that you present.
ReplyDeleteYes, good points, as usual, campanari. Thanks for sharing!
DeleteYeah, one reason was the strong focus on pitching pre-Barr with the top picks. Before him, almost every first round pick since Sabean's first draft (or two) were for pitchers. Since Barr joined, it has been almost alternating between position and pitching, and as noted, it takes a good number of years to see if there is success or failure.