Speed kills, or so it’s said. And given the prospects of this particular Blue Jays team, I was interested to see what kind of impact stolen bases might have on the outcome of a team’s season.
With that in mind, here are the top ten teams and the bottom ten teams in total stolen bases over the past decade (2002-2012) and their associated total WAR. (There's more here than just a table, dont worry).
Top Ten vs Bottom Ten
|Avg = 26.98|
|Avg = 21.78|
The premise of the investigation was to see if there was a relationship between the number of bases teams steal and the total team WAR. Presumably, if a relationship were found, this would indicate that teams stealing a lot of bases can expect to have a better team outcome. Here is the total data of all 332 teams between 2002-2012 in a more simplified form:
|*From 2002-2012||MLB Stolen Bases||MLB WAR||Number of Teams with more SB|
|Total bases stolen||31398||7146.6||n=||Avg WAR|
|Top Ten||156 --> 200||36.3||10||26.98|
Given what you see here, it clearly paints a picture that the percentiles of successfully generating stolen bases (SB) and the percentiles of successfully generating WAR are not correlated strongly as a team value. There is a small trend once teams get past the 75th percentile of SB upwards in team WAR. Is it significant? Not really. But is it inconsequential? Well... maybe I’ll leave that for you to decide. Baseball is still a sport where what counts in the end are wins and losses, and therefore any additional WAR generated is likely to help the prospects of a team succeeding over the course of a season. However, given the difference between the 97th percentile (top ten teams) and the median value of SB over the last decade works out to an average of 4.5 WAR, it’s evident that speed alone has little impact on the success of a team. The best thing that can be said is that great SB teams seem to do better than average or poor SB teams. However, great WAR teams are not dependent, or even particularly influenced, by SB values.
In short, speed alone is unlikely to propel a baseball team to glory. There is no apparent correlation between teams that generate a higher WAR value and teams that steal the most bases in terms of total WAR. This holds true by visual estimate (a gap of nearly 10 WAR between the best SB teams and best total WAR teams) and also if you generate a correlation coefficient across the sample (r=0.138). However, I’d wager the small benefits gained from the threat of a SB will help push good teams towards being great, as evidenced by the consistent increases noted from 75th percentile up, and the contrast in average WAR accumulated by top ten SB teams vs the bottom ten SB teams over the last decade. It’s simply not a requirement of being a good team.
End story right? I still wanted to look at this year’s roster since I was still curious how many bases they might steal. So despite the small returns in finding a correlation between the two values, I went ahead put together a Blue Jays roster similar to what we should see on opening day. If it happens to reach above that 75th percentile (or better yet into the top ten) of SB, at least it would be reasonable to assume there is more than just entertainment value associated with the Blue Jays running a lot this year. I used Bill James’ projections for each position player to project total SB (seemed like a reasonable way to assess probable outcomes without imparting any personal bias on the totals, and they were easy to access).
I included Henry Blanco as the reserve catcher because based on comments from AA recently, I’m of the assumption Josh Thole will be starting the season at AAA Buffalo. However, since the projections overestimate total games played, and I’d rather be on the low side than the high side of an estimated total SB, I cut the totals own by 6.4% to bring games played values back in line with a 162 game season from 9 position players.
With this adjusted value, 168 stolen bases puts the 2013 Blue Jays firmly within the top ten teams at stealing bases in the last decade. Does that mean we can expect a high WAR outcome from this Blue Jays team based on this data? No. Strictly, no. As discussed above there is no direct correlation between WAR values generated by a team and the number of bases they steal in a season. What we can say from this data is that the Blue Jays are more likely to produce a higher WAR by stealing a lot of bases than by not stealing a lot of bases. And it’ll be fun to watch.
So it’s a simple, if slightly disappointing, outcome that suggests:
- Stole bases are worth more than just entertainment value.
- They give a slight advantage over other teams that do not steal bases, but
- They will not directly influence a team’s performance in generating a high WAR value.
Disclosure and things I have assumed: I have assumed team WAR is a reasonable predictor of team success, and I have assumed total SB generation is a reasonable predictor of successful baserunning.