Phillies' recent Draft history: Boom and Bust

We've seen the recent reports about how bad the Phillies have been at drafting lately.

For example Todd Zolecki broke it down this past summer in advance of the June draft - the Phillies haven't just been last in the value attained by the players they chose, they've even been far behind the second worst team:

According to Baseball Reference, the combined WAR (Wins Above Replacement) of Phillies draft picks over the past 10 years is 20.7, which is a remarkable 24.6 points lower than the 29th ranked Blue Jays (45.3).

Now, WAR produced by draft picks is not the only way to determine drafting success, though it is a convenient metric of the things that we can measure to one degree or another. And even if you do use WAR, total WAR produced by all of a team's draft picks isn't the only way to use it. You could instead count the number of players who made at least some minimal impact (say at least 5 WAR over their career), or a bigger impact (20+ WAR?), and so on. Or you could focus on just the WAR generated while they were still controllable, or also include prospects who reached a certain level of value (MLB top 100 prospects lists, or a team's top 10, etc.) and were therefore useful trade chips. The list could go on at some length.

However having said that, total career WAR is what we will use for this analysis, and we will look at it two ways: total WAR produced to date, as well as total WAR produced as a % of the total career WAR that would be expected based on draft position.

I had long been of the opinion that the woeful stat quoted by Todd was explained in part by having fewer picks and drafting lower due to winning records. It turns out there's some truth to that, but not nearly enough.

Expected Value from Drafts

There's been some work done on how much value (i.e. how much WAR) can be expected from a draft pick based on the round and draft position of that pick. One such model was developed by Sky Andrechek:

Using that model, we can calculate the total WAR a team's picks in any given year would be expected to accumulate over their careers.

Below is a graph of those values for the drafts since 1990. Some things to note on that graph:

- The value of a team's picks in a typical draft ranges from a low of ~17 career WAR over their combined careers (for teams that have forfeited picks and/or are picking late in each round), to a high of ~40 WAR (for teams that have supplemental picks and/or are picking near the start of each round).

- The decline in the MLB average expected WAR over time is a function of 1) the drop in the number of rounds from 80+ in the 1990s, to 50 more recently, and then 40 under the current CBA, and 2) the model, which continues assuming some minimal average value in the last rounds which is small and continually diminishing but adds some to the total.

- The spike in the expected value of Phillies' picks in 1998 was due to having the #1 pick in the draft (used for Pat Burrell) and picking first in each round, as well as a supplemental pick at #42 (Eric Valent).

What jumps out though is that for most of the past 15 years the number and draft order of the Phillies' picks have placed them below the average in the total expected WAR from those picks. In fact, from 2000 through 2014, they rank 29th in total expected career WAR. That's mostly a result of two factors:
1) drafting in the latter half of each round, due to winning records
and 2) ranking dead last in the number of first round picks in 2000-2013, as detailed here.

By the way, in both 2013 (picking 16th), and particularly 2014 (7th), the total expected value of the Phillies' picks ranked lower than we might have expected based solely on their draft position because, unlike many teams, they didn't have any compensation picks for lost free agents.

It's worth repeating that the Expected WAR below is only a measure of draft position, based on the rounds and draft order of a team's picks. It does not consider the players selected.

Legend: the red line is the total combined expected career WAR for each year's Phillies picks based on round and draft order. Green and yellow lines show the lowest and highest team totals in each year, and black is the average.

Actual Value from Drafts

The above graph shows the Phillies have gone into drafts for most of the past 15 years already at a disadvantage before ever making a selection. If their draft results had matched their draft position, they would have been 29th in total value over that time, and there would have been no five-peat, no continued success (tm), and no World Series title. Fortunately, they beat those odds in the early part of the 2000s and got more value than would have been expected.

Not-so-fortunately, they have not drafted well in recent years, even when considering their lower expected value. Below is the actual (Fangraphs) WAR that the players drafted in each year have accumulated so far in their careers (regardless of what team they ended up doing that with).

Note the sharp drop below the average line starting with the 2004 draft:

And below is the actual WAR again (the red line in the above graph), this time breaking out the biggest contributors. The light green bars represent all others drafted in that year, and the more significant of those are listed at the bottom.

The last graph of actual WAR shows the cumulative WAR compiled to date. For example, in all the drafts since 2000, the Phillies have selected players who have accumulated 188 WAR so far -- not close to the high of 262, but above the average of 163:

Expected WAR vs. Actual (so far)

Finally, below is the total WAR accumulated to date for each year's picks as a % of the WAR they would have been expected to compile over their entire careers. For example, in 2003, based on the number and draft order of the Phils' picks that year, those players would typically be expected to accumulate a combined 18 WAR between them over their entire careers. Happily though, one of those picks was Michael Bourn, who has already accumulated 22 WAR all by himself. Two other picks that year were Kyle Kendrick and Brad Ziegler, who have contributed another 9 for a total of 31, or 169% of what would be expected, well above the average.

That's the end of the good times though, as far as the draft is concerned. 2004 may still turn out to be average-ish if J.A. Happ hangs on for a few years, or if Greg Golson manages to make his way back to the majors and contribute. After that, the only relative bright spot is the 2008 draft, where Vance Worley and Jarred Cosart have so far combined to provide slightly more value than the average team's draft from that year.

Starting at the far left of the graph, the early 1990s were largely a disaster, with the lone shining exception of Scott Rolen in 1993.

However Jimmy Rollins' signing in 1996 signaled the start of one of the most successful periods of drafting success of any team in recent history. Even when including the below-average 1995 in order to get a full decade, in the 10 years from '95 to 2004 no team in baseball drafted better than the Phillies. The expected value from the Phillies' draft slots over those ten years was average, or about 254 total career WAR, ranking 16th in MLB. However they drafted players who have already compiled 351 WAR, which ranks #1 over that period (see table below), as does the actual value as a percentage of expected.

The effect of that success is still measurable today. As the graph below shows, if we take together all drafts back to the early 1990s for each team, the Phillies are near the top in draft value over those 20+ years. Even if we limit it to just the 2000s, the Phillies have been well above average in draft performance in 2000-2014, thanks to the success in that 2000-2003 period:

And while the focus here is WAR, as has often been noted the lack of production on the field by the more recent drafts isn't the whole story. The products of those drafts proved to be very valuable -- they were used to acquire Cliff Lee, Roy Halladay, and Roy Oswalt, and so were a key reason for the Phillies' success in 2009-2011:

Lee: Lou Marson (2004), Jason Donald (2006), Jason Knapp (2008)
Halladay: Kyle Drabek (2006), Travis d'Arnaud and Michael Taylor (2007)
Oswalt: J.A. Happ (2004), Anthony Gose (2008)

And as with Happ, someone like d'Arnaud could still have a nice career and make that year's draft a success in terms of WAR as well.

More recently, while there are still players from the 2009-2012 drafts who could pleasantly surprise, draft picks from those drafts have not only generally not done much yet in the majors (as might be expected in that short a time), but they also haven't produced many prospects who are projected to make an impact. Some of this may have to do with picking more high schoolers (until this year), who require more time to develop and make an impact, but I'll let the prospect experts speak to that.

In any case, as referenced above, below are the numbers by team for the past ten years, and for the ten years before that.

Expected WAR, and Actual WAR to date, by Team:

Below are two other summaries. The first is for 2005-2008, when Mike Arbuckle proved he was human, and the second is for the entire period since 2000:

Summary

By signing free agents and losing compensatory draft picks (and getting fewer picks by that means), and more recently because their winning records put them in the bottom half of the draft order, the Phillies have had among the worst draft positions in MLB over the past 15 years.

Despite that, they had enough success in the decade from 1995 to 2004 that they were arguably the most successful team in drafting over that period.

However that lousy draft position caught up with them in spades more recently, and in 2005 through 2014 overall, they've been the least successful team in MLB in terms of value generated so far by those picks.

Could JP Crawford, Aaron Nola, and the rest of the recent picks be the beginnings of another reversal in the team's drafting fortunes?

We can only hope.