The Difficulties Finding Good Players in the First Round of the Draft

I keep on seeing the comments about the Sabean era Giants being unable to evaluate talent.

So the below post is a comment I made and then added more.  I started it before Zaidi was hired, then just finished it up, since the point is moot now that Farhan and his new hires are in charge of the draft.

ogc thoughts

It is really difficult to find good players in the draft.  I've started my data collection and have some initial analysis which I'll use here and going forward.  Here, I'll discuss just how hard it is just to find a good player with a first round draft, especially when the team is playoff competitive most of the time, and really, any time you are not tanking your team.

I would argue that there is no fan who can tell whether any particular GM is bad or not. The odds of finding a good starting player is extremely low, even for the top 5 picks overall, even for the first pick overall, where the success rate is still under a coin flip (my stats say 49%).

By the end of the first round, which is where the Giants have been mostly selecting from 2010-2017, the success rate is under 10% (roughly 7%; even their top picks were only around 15%, for the Beede and Bickford picks).  And when the success rate is that low, 56% of the time, you will see the team fail to find a good player via the draft over a 8 year (2010-2017) period, so it is not unusual to see a team not find anyone in the first round, even over such an extended period. So, while many teams find a good player during that period, it is not that unusual either to see a lack of success, based on random processes.  And given how good Ramos is, there's still hope that one might develop, with time.

Not Only is Finding Them Hard, It Takes A Lot of Time To Develop Them

And that's just overall, when all is said and done. Then you got the 4-6 years it takes most prospects to reach the majors when you are that far back in the draft, particularly for HS prospects. So, for example, at the time that Ramos was drafted, in 2017 draft, 60% of the time, a playoff competitive team would have ended up with no good player over the prior 7 seasons, from their first round picks, but let's say he does become that good player, since he was still 17 YO for the purposes of that draft, he probably won't make the majors until 2022 probably, when its his 22 YO season, but it won't be unusual either for him to finally make it in 2025-6 when he's 25-26, so it is possible we won't know until 2025-6 that the Giants 2010-2017 picks were really that bad, or just random luck when the odds are so low for finding a good player, when a team is competitive.

And it could still happen for other first rounders. Stratton is one of those first round picks, and while he's old and he has struggled, he's also been very very good when he's on (he had a low 2 ERA over the 14 starts from end of 2017 to start of 2018), and Vogelsong reportedly taught him his proper mechanics while Stratton was in the minors after his latest struggles. Bumgarner had similar struggles until Tidrow taught him his proper mechanics, which he then could replicate over and over again.  And Stratton's curveball spinrate was elite in 2017 and 2018, he was in the 98th percentile last season.  His fastball was too, in the 91st percentile.  But his results were night and day between 2017 and 2018.  He could be very good if given the chance, but he needs to figure out what he was doing right in 2017 (or rather, the Giants coaches and analysts should be finding this out).

Thus, the book is not closed yet on Stratton. He could still turn out to be a good player, he was on a 4-WAR pace in 2017, if he can string together 4 seasons, covering 28-31 (2019-2022), that would put him into the conversation for being a good player selection.

Of course, easier said than done, but what I'm trying to illustrate is how hard it is to actually come out and say definitively that any GM is bad at finding good prospects, even after nearly ten years, when the odds are so low to begin with, and players ebb and flow in reaching the majors.

Illustration:  Histograms of Some Picks

Here is the histogram of the first pick overall (broken up into intervals of 18 bWAR, which is my standard for a good player):

Hard to see, but most first picks overall do not become good players.  Only 48.9% from 1965 to 2014 have become a good player.  That's quite a different picture from the mean of 22.3 bWAR, which suggests that you will find a pretty good player on average.  But as you can see, there are a number of outliers far to the right, which brings up the average greatly.  Still, the median is still 17.8 bWAR, which still is pretty good.  But the problem is that 51.1% ends up being in the first two columns, players who are bad or just okay, useful, but not the good star players you hope for when you say "first round pick".

Here it is for the second pick overall, same breaks:

Similar profile, but notice how the column to the right of 18 is much smaller than above.  Mean is now 15.3 bWAR, median 11.7 bWAR.  And only 29.8% of the picks from 1965 to 2014 became good.  That's bad odds against Joey Bart being a good, let alone great player. 

Here is the one for picks 21-30:

Now the players who are not good are the vast majority of the picks.  Only 7.0% are good players.  Mean of 6.8 bWAR, median of 0.7 bWAR.

Illustration:  Playoff Competitive Team Picks 10th Eight Seasons in a Row

For teams that are Playoff Competitive, they pick in the 21-30 overall pick draft range, and hence my last histogram.  So when you are playoff competitive, the hit rate on good players being selected drops to 7.0%.  Over a 10 year competitive period, assuming picks were done randomly, 48% of the time, you end up with finding no good player in the first round, or roughly half the time, it is not unusual for a GM to end up with nothing good found.  37% of the time, you find one good player, and 15% of the time, multiple good players. 

And that's by random luck, with no assumption on whether the GM is good or bad.  Meaning a neutral GM, neither good nor bad, would miss half the time over a 10 year period, find at least one half the time too.  It would not be very usual to see either happen. And random luck could end up finding multiple (but the luck involved with finding 3 or more is huge, only 3%).

Now, let's look at an 8 year period.  With those odds, the binomial probability over 8 seasons of being good, and drafting somewhere in that range, lays the odds at 56% that the team ends up without the good player the team hoped for when they were drafting in the first round.  34% of the time, they end up with one good player found, 10% of the time with multiple good players.  Again, bad performance expected most of the time.

Criticism of the Giants 2011-2017 Drafts

The Giants 2011-17 period has been put down by most as a poor job of selecting talent, but as the odds above shows, randomness and the odds suggest that the Giants were fighting bad odds, as around half the time, you end up with no good player drafted in the first round (as a proxy for how well a team does in the draft).  And 90% of the time, a GM can be expected to either find no good player or one good player.  And that's basically where the Giants are right now, with Ramos looking like that good player, though he has a lot of development yet to come (and Stratton as a potential one). 

On top of that, players take 4-6 seasons typically to figure things out and make the majors permanently (if they are good), so there is no real way of fully assessing even such a long period as 2011-17 until we get out to 2023, and get a much better idea of what the Giants received.   Right now, does not look like much, but by 2023, maybe Stratton, Suarez, Mac, Ramos all look like good players, proving the Sabean Naysayers wrong in their assessment that the 2011-2017 period was bad drafting.  The development of good players can be achingly slow oftentimes.

And maybe none of them push forward and be good players.  But that's not the point, the point is that we don't know RIGHT NOW for certain that the draft was fallow, we just know there isn't anyone good, yet.  There's this guy on The Athletic who loves to blast Sabean, and he don't care if he's wrong a year from now, and don't understand why I harp on him about that.  But what if he's wrong, and the firing just cost us big time, in terms of someone who can discern talent (as shown by Sabean from 1997 to 2010).  But without the ability to run different scenarios, we can't see exactly.  Or can we?

Because we had this scenario before, we can see.  Go back to the future, to after the 2005 season, and there was just as many of these John Does who hated Sabean and would berate me when I celebrated Sabean's two year extensions he was getting back then.   These people would have fired Sabean, and with him and his team gone, and a new regime following new methodologies unlike their predecessors, as the assumption here is that the old was bad.

Thus, if this had happened, there is a very good chance that there's no Lincecum drafted, as the consensus was that Lincecum was going to be a reliever, and nobody wastes a first round pick on a reliever or on anyone as short as he was.  There's a significant chance that there's no Bumgarner, as most did not like his cross body throwing, and he didn't even start throwing non-fast balls until late.  Keith Law, a draft expert who has worked in the MLB, mocked him at #28, which is around where we picked up Wendell Fairley (whom Law correctly mocked!) but there were others that had him coming in the early teens as well, opinion was split on his value, but it was Tidrow who saw his tremendous value (when interviewed right after the pick, the Giants noted that they expected Bumgarner, a high school pitcher, to reach the majors in two years; they have never said anything similar before or since). 

And there's no 3 in 5 without them, we needed both.  Daniel Bard was mocked to the Giants when Lincecum was selected; Beau Mills was the hitter the Giants was expected to select when they got Bumgarner, since many thought the Giants needed hitters.  Neither developed to be good players, Mills never even made the majors.

Sabean Era Has Been Good

Thus, I still believe in the Sabean who had been so good at identifying who to keep and who to trade (really great, he only let go of Liriano and Foulke, but kept Cain, Lincecum, Bumgarner, Sanchez, Posey, Belt, Crawford, as well as finding other players who contributed significantly, Sanchez, Wilson, Romo, Panik, Duffy, others), but even I can't say for sure that he hasn't regressed. And I would be first in line to push him out the door if I found facts to support that stance. But with Ramos and Bart selected, they look like very possibly to be the next good players found by him and his front office, but it will take some time to find that out.

Even Bart, #2 pick overall, so far, historically, from 1965-2014, as noted above, only 30% of them ended up good, so that means 70% of the time, you end up with less. If you lower the bar to useful players, like a Joe Panik, the odds are still under 50% of finding at least a useful player, half the time you end up with so-so players (or career minor leaguers).  And we're still not even sure if Panik is a good player or a useful player, he started out good, but the past two seasons make him look more like Keith Law's prediction of a utility player.  But he could swing things back with a couple of good seasons.

Sabean Era is Now a Retrospective, It's Zaidi Time

In any case, it is a bit moot now, since Zaidi has now taken over, and we have new personnel in place to handle the draft.  This is now a historical examination, as well as an illustration of the difficulties of how hard it is to say whether a draft period has been bad or good, even over an 8 year period.

Someone made the point that it has to be better than what the Giants did under Sabean since the great Belt draft, 2010-2017.  But as I noted above,  the same could be said for his drafts from 1997 to 2005, then, as well.  If he were fired then, there's no 3 in 5, probably, Lincecum and Bumgarner were not the obvious picks, neither was Crawford or Belt, and Sanchez was a wild shot in the dark.  So if it was wrong then, it does not mean that it was right now.

Is it random luck?  It seems to be, with my illustrations above of historical rates of finding good players, and how hard that is when you are competitive, but not totally sure. 

But the people saying that Sabean (and Barr) are done don't really have any hard facts supporting them, either.  The lack of finding good players is not evidence (see 1997-2005) that they weren't doing their job, it is a sign of the very very very bad odds that happens in the MLB draft.  And the odds of selecting Lincecum, Bumgarner, Posey in a row, randomly, as I investigated in another post, was sky high, 1,000 to 1 odds, from what I remember.

As I showed above, it is too soon to say that the 2010-2017 period was bad, with Stratton and Ramos still interesting prospects, in terms of becoming good players, plus any success from Suarez, Mac, and Duggar, could rewrite the draft story relatively quickly, with a good 2019-2020 from this group.  The perspective of how good or bad the draft period is seen can swing widely if they end up being good or bad, much like the view of Panik's projections.

All in all, I think Sabean and gang have been good talent evaluators, overall.  Did they fail us from 2010-2017?  I say it's hard to say, definitively, given the odds against being good in selection with the draft, and yet, they did need to do better.  But there are still eggs yet to be hatched in Stratton, Suarez, Duggar, even Shaw and Mac, plus Ramos and Bart, where the odds were still against them finding a good player.  And I feel good enough that ultimately, the end of the Sabean era will confirm that they were still good talent evaluators, that it is just the poor odds that makes things look bad.  But the clock is ticking away.

And now that we have Zaidi, and Sabean is his Yoda, I am hoping the Giants can be even better.