Farm Systems Are Cyclical
Everything in life is cyclical; just look it up. Depending on your subject of interest, you can likely find some meditation on the cyclic nature of events in that particular field. Politics, history, sociology — heck, even opera has cycles. Baseball is no different. Whether it’s the dynamic between pitchers and hitters, strategic trends, or the success of individual teams, each side of the coin will come up eventually.
One key area in baseball that exhibits this cyclic nature is in the relative strength of organizational prospect talent. Every year, we see various outlets publish their rankings of the league’s assorted farm systems, thus giving hope to downtrodden franchises everywhere. Baseball America just released their list, while Baseball Prospectus and John Sickels will surely follow soon with their own. As all these lists are published, it’s worthwhile to consider how clubs tend to navigate the crests and troughs of such rankings, and what that movement implies for a team’s future.
Organizations or specific GMs tend to become aligned with a particular philosophy of how best to utilize a farm system. Dave Dombrowski, for example, has exhibited a tendency to trade prospects for established stars. The rankings illustrate this tendency. Looking at Baseball America’s organizational talent rankings from 2001 to 2018, one finds that Dombrowski had a bottom-10 farm system 10 times in his 14 years with the Tigers. Meanwhile, a small-market team like the Rays has a reputation for cultivating a collection of prospects who can help at the major-league level. The data supports this, as well: according to BA’s rankings, Tampa Bay has featured a top-five system seven times in the past 18 years.
Even teams with established tendencies, however, still have their relative ups and downs. The Rays spent 2014-2017 as a mid-tier system ranked in the teens. Dombrowski had a couple of years early in his tenure with Detroit during which the system ranked in the top half of baseball. Even when a farm system is bad, there will be upticks. The Tigers’ 2012 farm ranking (No. 23) might seem pretty bad, but it was still a peak in rankings in between the troughs of 2009 (No. 28) and 2015 (No. 30).
So what can this cyclic pattern tell us? The first question we need to consider is how long it takes a club to travel from the top to the bottom of the rankings, and vice versa. To get an idea of this, we first need to define what a “peak” and “trough” are in this context. The Arizona Diamondbacks provide a useful example, in this case. Consider their rankings since 2001.

There are two clear peaks and troughs, with one possible mini-cycle in 2014-15. However, for the purposes of these measures, we will only be counting the two clear peaks and two clear troughs. For some years and teams (the 2006-12 New York Mets, for example example), there are no clear peaks or troughs, with alternating rises and falls each year.
On average, it takes 2.8 years to go from the end of a peak to the end of the trough. It takes almost an identical amount of time (2.75 years) to go from the end of a trough to the end of the peak. However, the distribution of both peak-to-trough and trough-to-peak movement is skewed out to the right, with a few long rises and slow drops. The most common wait until the worm turns (in either direction) is two years.

The next question is how far teams rise and fall in their journey from peak to trough and trough to peak. Intuitively, one would suspect that a longer duration between extremes would equate to a greater rise or fall in the ranks. This is reflected in the data. What we can also see is how symmetrical the peak-to-trough and trough-to-peak are in the relationship, having nearly identical rates of change given the length of the decline/rise.

On average, teams experience a trough-to-peak change of 17.8 spots upward in the rankings, while a peak-to-trough change averages 18.1 spots downward. The precise numbers are highly dependent on how long a club is falling or rising, of course, with longer runs resulting in greater changes.
| Years | Peak-to-Trough Fall | Trough-to-Peak Rise |
|---|---|---|
| 1 | 14.3 | 13.9 |
| 2 | 16.4 | 16.1 |
| 3 | 18.5 | 18.4 |
| 4 | 20.6 | 20.6 |
| 5 | 22.7 | 22.8 |
| 6 | 24.8 | 25.0 |
| 7 | 26.9 | 27.2 |
Finally, we have to ask what happens at the end of these cycles. What happens in the years after an organization’s long steady rise — like what the Cubs experienced from 2009 through 2016? What about if that steady rise is followed by a quick descent — again, like the Cubs of recent years?
Unsurprisingly, the clubs that go through longer periods of rise see more wins on average in the three years following their peak. Periods of quick rises or quick drops produce similar results on average, while long steady falls through the rankings result in the worst average win totals.
| Type | Historic Examples | Current Incarnation | Avg W/162 (Next 3 Seasons) |
|---|---|---|---|
| Long Rise | Rays (2002-10), Pirates (2008-14) | Phillies (2012-2018) | 88.0 |
| Quick Fall | Athletics (2005-07), Orioles (2010-12) | Red Sox (2015-2018) | 81.3 |
| Quick Rise | Marlins (2009-11), Nationals (2010-12) | Padres (2016-18) | 80.6 |
| Long Fall | Braves (2003-07), Indians (2003-08) | Royals (2013-18) | 77.1 |
Looking at what preceded each of these changes shows the clear importance of a long run towards the top of the organizational ranks. That infusion of talent can help stave off brief downturns in a farm system and can even mitigate a longer run of lesser talent.
| Type | Preceeded By | Historic Example | Seasons | Avg W/Season (Next 3 Seasons) |
|---|---|---|---|---|
| Long Rise | Quick Fall | Dodgers (2011-16) | 2 | 97.5 |
| Long Rise | Long Fall | Cubs (2002-15) | 20 | 87.5 |
| Quick Fall | Long Rise | Cardinals (2005-10) | 9 | 84.7 |
| Long Fall | Long Rise | Giants (2006-14) | 17 | 80.1 |
| Quick Fall | Quick Rise | Milwaukee (2011-14) | 47 | 80.0 |
| Quick Rise | Quick Fall | Tigers (2003-06) | 50 | 79.2 |
| Quick Rise | Long Fall | Twins (2005-10) | 18 | 78.6 |
| Long Fall | Quick Rise | Angels 2010-16) | 21 | 73.8 |
The Dodgers, Astros, and Cubs all have created their success in part by extending the upward swing of their respective cycles. With this in mind, it’s easy to understand the excitment of the Phillies and White Sox, as they have also made a long consistent rise up the ranks. On the opposite end, the worries of the Orioles and Marlins, who are coming out of long downturns without long rises preceeding it, seem completely founded. But what all these fans need to remember is that, at some point, the second face of Janus will turn towards your team. The strong systems will descend into a trough and the weak will reach their peak, and the next cycle of talent will begin for every team.
Stephen Loftus is a Visiting Assistant Professor in Mathematical Sciences at Sweet Briar College in Virginia. In his spare time he usually can be found playing the pipe organ or working on his rambling sabermetric thoughts.
One thing I want to point out is that the ranks of farm systems are not linear in the sense that the gap between 1st and 6th place is generally much larger than the gap between 25th and 30th.
So the typical gain in value of jumping up 20 spots is going to be larger than the twice the typical gain in value of jumping up 10 spots.
“You’re in Carcosa now.”
Fantastic stuff. Is it even possible for the Red Sox to recover to a peak within 2-3 years at this point?
It seems this analysis supports the idea of tanking and a slow, methodical rebuild, much like Houston/Cubs did and the Phillies are trying now. Patience seems to pay off – stockpile a deep, talented farm to get enough “hits” at the big league level that you see successful results W/L-wise.
I know I’m being a smart ass here but how could it be different?
Good team is:
-drafting lower
– trading guys
-losing comp picks
– Maybe graduating some guys earlier (like the classic pitcher call up to relieve even if he is not quite ready to start like buehler last year)
Bad team is:
-drafting higher
– selling veterans for prospects
– Maybe getting some extra comp picks
That is why what the Dodgers and Yankees did was so impressive. Sure the braves,Astros and cubs did a great job but they also tanked big time. Ny and la acquired that farm while having winning seasons (Yankees did a small sell off while Dodgers won 90 games all the time). sure both of those teams did some cheating in the international market (or better using existing loopholes with boatloads of money) but still a great job.
Come on. The Yankees spent 220+ million a year to have 4 -80 something win teams from 2013-2016 without making to the LDS once. This is after having a top 5 farm system in 2011. Pretty much everyone contributing to them now (Judge, Sanchez and Severino) had been in the system as if 2012. The other top prospect in the system now were obtained selling off while within 5 games of the WC in 2016. Not a lot of teams have the financial clout to sell while that close. Miller and Chapman were used to to acquire many of these prospects , both acquisitions that other teams passed on since they didn’t have the money. McCann And Beltran, also assets acquired via FA obtained a nice bundle as well.
Also, they have been fortunate in trades, giving up very little for Chapman or Stanton who were dumped by tanking teams due to their salary
Not impressed. Just means the model is lacking a financial component
Just a nitpick but the Reds didn’t dump Chapman over salary concerns. It was because he choked his girlfriend and fired a gun into a wall to frighten her and the Yankees didn’t mind
To continue nitpicking, the Reds supposedly weren’t aware of Chapman’s incident until after the Reds and Dodgers had agreed on a trade iirc.
The Dodgers rebuilt the farm in three ways: they’ve always had strong talent evaluation, and they paired that with using their payroll space to acquire poor-value contracts for decent players (Adrian Gonzalez not worth his contract, but was a useful player) along with some prospect value.
They also have really developed players well (Bellinger is a 4th round pick; Seager was a late 1st round pick expected to move to 3b) and have also found buy-low opportunities (Walker Buehler is now their top prospect; he was a high 1st round talent but was actually injured when he was drafted, his first action as a professional baseball player was to get Tommy John .. so they could also get him several rounds later than his talent).
So they have used their financial advantage, but combined that with being really smart.
There are a lot of teams with financial advantages; the Cubs have not built a system that extends their window substantially. The Astros have currently done so. there’s now an expiration date on this great Cubs team that the Dodgers and Astros aren’t dealing with.
That was my thought when I saw the title…”Baseballs are round”.
“Sure both of them did some cheating in the international markets” and “impressive” don’t belong in the same sentence.
The Dodgers also exceeded the luxury tax to acquire several top tier young players. For example, to acquire Alex Wood and Jose Peraza, the Dodgers took on the contracts of Mat Latos (9M), Michael Morse (7.5M), Bronson Arroyo (9.5M).
They took on a total of $25M in dead contracts to acquire 2 top young players. I wouldn’t mind this maneuver if said team wasn’t exceeding the luxury tax threshold to accomplish this, but they spent $300MM that season. I don’t find rich teams exploiting the system to get better remotely impressive.
Fourier transform please, or access to the data?
Yes, team-by-team raw data would be appreciated here. A Fourier transform would also be fun to see, although I’m not sure it’s going to be informative for this particular body of data.
Nice piece.
BP released their Top 101 Prospects list four days ago: https://www.baseballprospectus.com/prospects/article/37535/baseball-prospectus-top-101-prospects-2018-top-mlb-prospects-ronald-acuna-victor-robles-vladimir-guerrero-jr-eloy-jimenez/
Is that a graphing oxymoron representing the trough as a peak, and the peak as a trough?
Does that mean the FO isn’t that important? It’s all about draft position anyway?
Cycles are the wrong concept. You’ve identified trends, the idea of repreating cycles, whether in farm systems or “Politics, history, sociology — heck, even opera…” is erroneous. While there is valuable information in the data you present, coming to the conclusion of “Everything in life is cyclical; just look it up” is intellectually lazy.
“Technology is cyclical.”
-The Beeper King.