Span vs. Morse, Base Running vs. Power

Most Nats fans have come around on the point that Denard Span is an upgrade
over Michael Morse. You replace a negative 20 runs in UZR/150 with a positive
4.6 and you upgrade the Nats weakest line-up spot from the past few season with
a career .357 OBP hitter. Defense is one thing but Morse was a left
fielder/first baseman and both of those positions are known for offense. At the
end of the day people are going to see the Nats replacing a career .839 OPS and
.363 wOBA hitter with a .746 OPS and .332 wOBA one. That ignores a large chunk
of offense and most of Denard Span’s added value outside of defense and OBP,
and that is base running.

Think about it for a second. What is the
first stat you look at when judging a player? Is it their slash line, wOBA,
WAR, what? How often is the first stat checked a base running or defensive
stat. WAR does try to account for those things and it does a good job of it,
but do most realize it or do they even care or are they even the ones I am
writing this for? I often even forget that the base running aspect of fWAR can
be viewed on Fangraphs. BSR/UBR is the stat and it takes everything into
account. It even counts for when a batter doesn’t run on a strong armed
outfielder. Take a look at the primer. Morse is a -10 BSR/150 base runner and Span a
6.3. As with defense, that is a large increase in value. From base running and
defense alone Span is worth four wins more than Morse.

This is all a little complex and may be
hard for some fans to get a grasp of. Stats should be simple so that everyone
can understand them. Converting base running to a runs number like UZR is nice
for those of us that cotton to that sort of thing, but what about for the fans
that don’t. Think about how easy wOBA is to explain. It is a formula that
values the type of hit and the times on base to a weight and is then converted
to be read like OBP. wOBA also tries to factor in base running, but only with
steals and caught stealing and in my estimation does a poor job of it. All the ways
of reaching base are given context weights of base outs situations while a
stolen base is weighed as 0.25 runs and a caught stealing 0.50 runs. Meaning
that Ian Desmond who stole 21 bases and was caught only 6 times for a 78%
success rate is gifted with 2.25 runs in the wOBA formula where a single
homerun counts as 2.058 runs. One single homerun is not almost as valuable as 21
stolen bases, in my opinion. And that is what we’re trying to measure when
comparing Span to Morse. Span is a speed player and Morse a power player. There
are no good stats that accurately measure power and speed and give us a way to
compare them.  

Making a simple stat that measures both
power and speed isn’t all that difficult. I am sure you know what SLG is. It is
the main stat by which power is judged and it is one of the simplest formulas
in baseball. It is total bases divided by at bats. Simple see. ‘Total bases’ is
a stat that measures bases that a batter batted themselves to. It isn’t too
difficult to go on ahead and add bases ran to into the total base stat and come
up with an offensive stat that measures both power and speed equally. For Morse
we will only look at his career with the Nationals as his power took a huge
uptick, and for Span we will look at his entire career. Also not all bases ran
to will be counted. Going from first to second on a single isn’t a product of
speed or any good base running. It is just playing baseball; the same with
going first to third on a double or first to home on a triple.  

When adding in stolen bases, bases taken
(bases advanced on a passed ball, wild pitch, sac fly, ect) , extra bases taken
(advancing more than one base on a single and more than two on a double) and
subtracting caught stealing and times picked off to career total bases Span’s
SLG upgrades to .503 and Morse’s .561. Morse’s power still gives him a sizable
advantage but it is much closer than their unconverted SLG of .389 for Span to
.514 for Morse. Now we can go ahead and add OBP to this to make an OPS type
stat that is .860 for Span and .904 for Morse in his time with the Nats. It is
closer than it was before and Morse still has the advantage, but Span is not
nearly the offensive downgrade his standard slash line would lead one to
believe.  

Simply adding more bases to total bases
may not be the overall best way to compare power to speed, but it is a way. It
gives us even more of an idea of what Span is going to add to the team
offensively and that the Nats aren’t sacrificing much if any offense for
defense. Consider for a second that the Nats lead-off hitters in 2012 had 762
plate appearances and reached base safely 32.5% of the time. Figuring that Span
will do so at his career average of 35.7% and will get around 650 of those
plate appearances he will reach base 232 times. If the Nats get an equal number
of PA from the lead-off spot in 2013 as 2012 and other Nats hitters perform to
last season’s average then the Nats are looking to go from 247 base runners
from the lead-off position to 268. Those 20 extra bases might not sound like a
lot, but they are 20 less outs and 20 more opportunities for either Werth or
Harper to bat. And if we count a base as a quarter of a run it is five extra
runs or half a win.

Add that all in to everything else we
have said Span is doing for the Nationals and he could be as much as a five win
upgrade over Morse due to defense, base running, and OBP from the lead-off
spot. That is a large leap for a team that won 98 games in 2012. Make of that
what you will. It is going to be interesting to see how all this plays out in
reality and if Span’s base running really does help to close the offensive gap
between he and Morse. 

Normal
0

false
false
false

EN-US
X-NONE
X-NONE

/* Style Definitions */
table.MsoNormalTable
{mso-style-name:”Table Normal”;
mso-tstyle-rowband-size:0;
mso-tstyle-colband-size:0;
mso-style-noshow:yes;
mso-style-priority:99;
mso-style-parent:””;
mso-padding-alt:0in 5.4pt 0in 5.4pt;
mso-para-margin-top:0in;
mso-para-margin-right:0in;
mso-para-margin-bottom:10.0pt;
mso-para-margin-left:0in;
line-height:115%;
mso-pagination:widow-orphan;
font-size:11.0pt;
font-family:”Calibri”,”sans-serif”;
mso-ascii-font-family:Calibri;
mso-ascii-theme-font:minor-latin;
mso-hansi-font-family:Calibri;
mso-hansi-theme-font:minor-latin;
mso-bidi-font-family:”Times New Roman”;
mso-bidi-theme-font:minor-bidi;}

6 comments

  1. I’ve long had in my mind a stat that would be easy for laymen to create and use to measure offensive contribution, my idea was for Absolute total bases, that would take TBs, BB, IBB, SB and xtra bases taken (the 1st to 3rd, 2nd to home on single or 1st to home on double) divided by PAs to give you a bases per appearance avg. You could then multiply that by 4/5 to give you a per game base expectation, or adjust the whole batting order for an bases per game expectation. I’d be very curious to see how many bases you’d average for run produced league wide…

    Like

    1. I actually did that at first for Morse vs. Span and did it on a 150 game basis, but lost the spreadsheet. Because of the extra walks Span ended up pretty far ahead of Morse if I remember correctly. I should figure out league average someday, or next time I randomly decide to create a stat.

      Like

      1. Oh please do, my other idea was to make a matrix of all possible outcomes from an at-bat including all men-on-base scenarios and assign a "value" to each and thus create a running value of contribution, you could even try to weight it by game situtation, but I’m (for the most part) opposed to that, I don’t believe in "situational hittng" as a tool…anyway, I’m not sure how to handle SB or taking the extra base (TTEB stat?) in such a matrix, maybe add it back into the batters previous outcome (ie single with SB would be "equal" to a double if the bases are empty..

        Like

      2. I have to admit I am confused by your comment since there does exist such a thing as a run expectancy matrix and it can be accessed for free for any year on Baseball Prospectus (http://www.baseballprospectus.com/sortable/index.php?cid=1122396), and I figured you would know this. I also would say I don’t know if situational hitting is a tool but not being a dumbass is. A dead pull hitter isn’t going to suddenly be able to go the other way when the situation calls for it, but someone like David Wright who is the best in baseball at hitting a ball where it is pitched will be able to. The research to really know if this is a skill demands access to both pitch/FX and hit/FX to see where pitchers pitch the ball in situational hitting situations and where batters place the ball. But again if a dead pull hitter is up in any situation the smart thing to do is to pitch him away regardless of the situation, but I do believe there are some hitters that are good at hitting where the ball is pitched and at the valuable skill of not being a dumbass, which is all situational hitting is.

        Like

      3. I completly agree with your take on situational hitting, I’m thinking more of clutch as defined by hitting better in high leverage situations. I think this is an illusion based on perception instead of statistics. There certainly can be such a thing as being a choker.Going deeper, I guy may seem "clutch" because he’s a great FB hitter and if it’s 2 out bases loaded in a tied game you can guess what he’s going to see

        Like

      4. As for run expectancy, I was referring first to finding a expected number of bases (based on the absolute total base stat i wanted to create) and what the "burn" rate would be, ie how many bases do you (on average) need to produce to get a run? I’d guess it’s around 7, but don’t have time to run numbers like that. The other was based on creating a point system to weight the outcomes and then compare that to runs actually produced to get a run expectation based on each batters "contribution score". This would all work better at a bar over beers, maybe Riggleman could join us at Champs? 🙂

        Like

Comments are closed.