Jeff Samardzija's switch to starter shows some old-fashioned outside the box thinking

Cubs fans have been celebrating for months now as they finally hired some forward-thinking executives.  Much fun was made of the old-ways of Jim Hendry and his staff, many times deservedly so, but those types of evaluations still have tremendous value, most notably when you’re talking about high school and international amateurs.

But can old subjective school scouting and development still play a role in finding a new market inefficiency?

I think it can, even if the previous statistical record suggests otherwise.  This is not to diminish the value of statistical data.  It is central to any team’s success to employ a more modern method of statistical analysis.  But considering almost every team now does it to some degree, has it ceased to become the unique outside the box thinking that once gave a handful of teams a huge advantage?

Statistical analysis is still a tremendous tool and is perhaps the biggest key to a Cubs resurgence, but I also wonder if the pendulum hasn’t swung just a little.

If I may make an analogy here, the way I look at statistical projections is the way you might look at betting on the horses.  Sure the 3 to 1 favorite is the most likely to win.  It’s the safest bet and given the choices and data, a win is the most likely outcome.  But most people have the same information, will bet on the same horses and in the end you will almost certainly lose more money than you win if you only bet that “most likely outcome” every time.

Sometimes the key to winning with horses can be knowing which longshot to bet on once in a while and make your big gains there.

One longshot the Cubs seem to be betting on this year is Jeff Samardzija as a starting pitcher.

What a pure statistical analysis/projection discounts is the possibility for individual improvement.  That is not because analysts think it doesn’t happen, but because it’s impossible to quantify with objective measurements.  It recognizes player improvements as larger, general trends but does not account for a skill or two that a player might develop on his own.

Let’s look at Bill James’ projections for Samardzija: (all in relief):



4.66 ERA

7.34/5.28 strikeout to walk ratio

It’s not encouraging and while there is some variance among other prognosticators, there isn’t anyone out there who sees him as being much more than just a replacement level pitcher.

But Bill James will be the first to admit he doesn’t know all the nuances of baseball on the field.  This is an excerpt from an excellent  interview he had with the guys at Obstructed View.

There are many areas of the game that I know nothing at all about… I know nothing about international baseball.  I know nothing or next to nothing about pitch calling (from the catcher’s standpoint).    I know next to nothing about what is called mechanics.    All that stuff about who lines up where on a relay throw; because I didn’t play the game at a high level, I don’t really understand that.    There’s a lot of stuff I don’t understand.

So given that gap between statistical analysis and the nuances of the game on the field, how do you use numbers to account for whatever mechanical adjustments Jeff Samardzija made to improve his command and secondary pitches over the offseason?  How do you account for a change in approach in terms of objective analysis?

The answer is you can’t.

I’ve talked to a few of my industry sources and they believe it’s absolutely possible for a player to make that kind of dramatic improvement practically overnight.  The odds aren’t necessarily geared toward that happening, but if it did, it certainly wouldn’t be the first time a player suddenly turned things around.  We once saw the Cubs own Ryan Dempster go from inconsistent reliever to a #2-3 type starter, though his case is a bit different because he had been a starting pitcher in the past.  He’d already had the repertoire of a starter, it just wasn’t needed in the bullpen.  To a lesser degree, we saw Matt Garza out-pitch expectation last season.  At the time, Fangraphs’ Dave Cameron  compared Garza with the decidedly mediocre Aaron Harang. Yet a change in his approach allowed Garza to trump even the most optimistic 2011 projections.  The same could be happening with Chris Volstad this season. Admittedly, though, a big year from Samardzija would top all of those cases and easily be the biggest surprise.

I’m not knocking advanced statistics by any means. I use them frequently in my own analysis and projections, as you all know.  What I’m saying here is that I’m not sure that’s the only way to think outside-the-box anymore.  Most teams have the ability to track and utilize modern statistical data now, so taken in isolation, those numbers don’t give teams the same kind of edge that it used to.

Maybe it’s not as much anymore about just identifying and following the right trends and playing the best odds. That will always be a part of modern baseball, but perhaps a new market inefficiency will be the ability to spot those players who may be just an adjustment or two away from bucking those odds.  It’s more art than science and it takes combining the right player with the right scouting, instruction, and development.  Only time will tell if Samardzija will be that player this season.



Leave a comment
  • fb_avatar

    This article pretty much lays out the outline as to why I love this article. Modern ways of looking at the game but not forgetting every little thing that can change the outcome on the field that number crunchers know nothing about. Which is why I don't read the Bulls bashers like Hollinger on espn

  • In reply to Nate Hummel:

    Thanks Nate! I have to admit I'm something of an analytical thinker (was a science teacher a long time ago), but I also appreciate the artistic side of things. To me, scouting is the artistic side of baseball and in many ways, it's the part of the game that fascinates me the most. One of the things I love about the game is that you can keep learning but you can never really know everything. That excerpt from Bill James, who's as smart as they come, is an eye opener and exemplifies why you should always consider the art of scouting along with the science of numbers.

    There's nothing that objectively tells us that Samardzija will be a success as a starter -- but that's what makes things so fun. I'd love to see him defy expectations!

  • fb_avatar

    Great article John. Some times a little gambling to go along with strict analysis is alright. The have little to lose in seeing if samarzdija can make it as a starter. now if were talking about a 4/5 starter i say keep him in the setup role. But if we can get a potential #2/3 out of this you'd be a fool in a rebuilding type of year not to see what you have. Couple that with the recent trend of power arm prospects going to the bullpen first to adjust to major league hitters and help the team until their time comes to start, look at Daniel Bard, Phil Hughes (who won 18 games the year after setting up for the yanks), Chris Sale, and most notably Pedro Feliz.

    Shark is a starter and is finally going to be one. I felt they should have just taken their bumps with him a while back after he got one spot start and threw 5 scoreless innings only to be sent back and forth from Iowa, if anybody remembers.
    He always had the stuff. power fastball with movement, plus slider that only gets nastier, developing secondary pitches and now improved command on a big strong body, and a competitor. Thats a number 2 starter if all goes well. I like to use a term they say alot at MLB network, "This guy is what they look like John" Ill put my bet that Shark has a good, randy wells typeish year in his first starting and then takes off the year after.

  • In reply to Marcel Jenkins:

    Well said, Marcel! Great examples of other teams trying something similar too.

    I do think it's worth it to at least gamble partly because there really isn't much downside. Worst case scenario is that he pulls a James Russell and winds up back in the pen. As we saw with Russell, it didn't hurt his effectiveness at all once he got back into his relief role.

  • fb_avatar
    In reply to John Arguello:

    My point exactly, you have to know Shark noticed James Russell was given a chance to start and it was terrible but he was still given a chance so their is no way to justify arguably a better overall pitcher with a much higher ceiling not being given a chance. like shark said, if he's not tried as a starter he'll need a good explanation why and I agree with him. his best argument is James Russell and in a way Casey Coleman getting the starts he did, Shark could have gotten those starts.

  • Some guys have all the talent in the world but are just knuckleheads.
    How do you measure a knucklehead? By the number of water coolers they demolish?

    Same goes for killer instinct and chokers. It's hard to measure such things...

  • In reply to eaton53:

    That is very true. When Shiraz Rehman was talking about players making individual improvements at the Bloomberg presentation, he really emphasized that it very much depends on the player and his personality.

  • Actually you can come to the conclusion that Samardzija has made great strides using statistical analysis as well. Despite the fact that the projections are seemingly insane doesn't mean a closer look can't bring you the same kind of results.

    Samardzija's walk rate plummeted as last year moved on. Something to the tune of:

    That 9.8%, where he finished up, is right in line with guys Tim Lincecum and Ubaldo Jiminez and even a tick lower than Gio Gonzalez and a couple of others. Not only that, but taking out his first 7 games in April, where he walked 13 men in 11 innings, he had a perfectly acceptable 4.2BB/9 in 77 innings.

  • In reply to doogolas:

    For the record I'm not saying that scouting doesn't play a role here. All I'm saying is that I firmly believe that statistical analysis played a part in F7 getting a shot this spring.

  • In reply to doogolas:

    Honestly if it were just a pure statistical analysis, I don't think he sniffs a shot at being a starter. This decision to me was based on more subjective criteria.

    Volstad is more of a guy to me with stats that indicate he could bounce back. He's also doing some other things, the grip and approach, that could help.

  • fb_avatar
    In reply to doogolas:

    great points there

  • In reply to doogolas:

    Oh absolutely. He's headed in the right direction, especially after an atrocious 2009 and 2010. I don't ever want to imply that stats aren't part of the equation. They're a big part. Was more emphasizing the things that are not as easily measured. I'm thinking if you went purely by statistics, you probably wouldn't recommend making Samardzija a starter. That could ultimately prove to be true, but I'll be pretty interested to watch.

  • In reply to doogolas:

    It's also right in line with James McDonald, Dillon Gee, and Charlie Morton.

  • In reply to GW22:

    None of whom strike out guys at the rate Shark does. That's why they weren't even worth talking about.

  • In reply to doogolas:

    Actually, they do. You must consider starters vs. relievers when looking at the numbers. 9-10% is not a particularly good walk rate. The guys you mentioned succeed in spite of that because of their overwhelming stuff.

    McDonald or Gee would be a pretty favorable outcome for F7, considering his track record.

  • Spellchecks record will be:

    8 wins
    3 losses
    3.97 era


  • In reply to johnnywest333:

    Not too shabby.

  • Most of us fit into the normal range of statistics - here is what usually happens. But, what the statistics don't account for are the intangibles of human beings.
    What happens when someone finds that passion, the heart to go for it, the courage to defy being normal? They move beyond the statistical analysis.
    I see it in my work with teenagers and it's fascinating when it plays out in sports. Samardzija - "Spellcheck" that's funny - seems to have found some of that passion, heart, commitment, and I am hopeful as a Cubs fan and compelled as a person to see how this plays out.
    And John, it may well be that the new outside the box, is learning how to see and understand when an individual is breaking out of normal and finding the courage, heart, and confidence to transcend what's expected.

  • In reply to Morgzie:

    That is something I think is often underestimated. It's certainly better to have just talent than just heart, but of course you'd like players to have both if you can.

  • fb_avatar

    The real question is justhow short of a rope they will give him. I hope management doesn't jerk him around like they have in the past.

  • In reply to Rich Cap:

    Good point, I hope they give him a legit shot. If he absolutely bombs that's one thing, but they should have some sort of commitment if they're going to do this and be prepared to live with the ups and downs.

  • fb_avatar

    Nice article John. Like the analogy with the betting on horses and I totally agree with you. It is the same approach I use when entering the espn tournament challenge, although I probably go overboard with picking the upsets. It may be safer to pick the favorite, but how does that distinguish you from the other 99% doing the same thing? The key is to have some way of picking the right dark horses. Obviously, if you can do that you will be golden. Not only that, but it makes for a much better story as well!

  • In reply to Gary Kueper:

    Thanks! I guess it remains to be seen if Samardzija is the right dark horse to bet on, but man, it would be great if the Cubs could find a solid starter within their own bullpen.

  • Hey John on another note with the way that Mather is playing, wouldn't you think that he would be a better solution in RF then DeJesus? We need power on the corners and Mather has that and from what I can tell he is pretty darn good in the field too. I don't know much about the guy but it seems to me like he is a much better fit. Am I wrong in this thought?

  • In reply to johnnywest333:

    You don't want to make too much of one spring. You're talking about replacing a player who has been an above average player in the OF his whole career for a guy who's basically been a career minor leaguer on the basis of 30 ABs in the spring.

    Mather's best value is as a guy off the bench who can play multiple positions for now.

  • Thanks John, I agree I just don't know enough about Mather, however he is lookin pretty darn good so far....

  • In reply to johnnywest333:

    I'm sure Sveum will find a way to get him into a few games early on while he's playing well. You can never say never, but for now you want to give the vets the benefit of the doubt.

  • I remain skeptical of his ability until he can prove it between the white lines this season. I would be happy if can stay healthy for the entire season and not pitch himself out of the rotation.

    Still a ways to go from where he is today to being a productive starter over a 162 game season.

  • In reply to JK1969:

    He doesn't have much of a track record, so I think it's natural to feel skeptical. I'm cautiously optimistic myself. Not going to get too excited until he shows he can do it for a greater number of innings.

  • Another question John, where do you see Amezaga ending up this year, he's tearing it up.....

  • In reply to johnnywest333:

    I'm a fan of Amezaga. I think he's a good little player with a high baseball IQ and enough athleticism to play all over the field. It seems to me he could be the emergency SS at AAA. With Barney as the team's current backup SS, they'll need him in Iowa. Tolbert also plays SS, but Amezaga is the better player.

  • fb_avatar

    Marmol out of game early. News?

  • I just did a quick recap, as johnnywest said, it looks like nothing major, just a cramp.

  • Patrick Mooney reports that Marmol felt a cramp in his right hand and will get an MRI as a precaution. He didnt seem concerned

  • fb_avatar

    John, thanks for the link back to Obstructed View. Great work here.

    FWIW, I don't think any statistical analyst would say that stats answer the entire question. I've been reading them for many years now and for a long time they've said it's half the picture. As tangotiger likes to point out, look at stats and scouting as different lenses. To get the best picture of a player you combine the two. The larger the sample size the less valuable the scouting is and vice versa.

    I wouldn't really call this decision outside the box thinking. It's something organizations have done for years including Jim Hendry. There's also no doubt that those you spoke to are right in that dramatic improvement happens, but I'd ask those people how often they've been right about which players they thought improved? We hear it every year. Scout A talks about changes to someone's game and then we see the same player. Every once in awhile there is a change, but most of the time the scout was simply wrong.

    This doesn't mean you ignore scouting. You can't. It's as valuable as stats, but the information is only valuable when it's detailed information. A scout saying "I think this guy has turned a corner" isn't detailed information. It's an opinion and it deserves some weight, but why has he turned a corner? Objective analysis includes some subjective opinions, but as few of them as possible. What is it about Samardzija that has improved? Is his fastball a mph faster and showing some movement? Is his slider biting more? Do the batters appear to be more confused by what he's throwing? There are many other questions of course, but it's that detailed information that is added to the mountain of statistical information to create one picture of the player.

    If you combine Scout A's opinion about this you may as well include Statistician A's opinion about it too and then you end up with an old-school front office making decisions based entirely on opinion rather than objective analysis.

  • In reply to dmick89:

    Thanks, dmick. And I agree with everything you've said here. We look at both scouting and stats here and I think that's implied throughout the article.

    But I do think this particular decision came was inspired from the scouting perspective. His improved command, his more consistent velocity, etc. are scouting tools, though they can be measured more objectively now than they were in the past. But I don't see this as the Cubs looking at his past numbers and seeing something that indicates he's ready to break out as a starting pitcher.

    The people I've talked to think it's possible for a dramatic improvement, although as I mentioned, it's something of a longshot, but perhaps it's one worth taking since the downside seems pretty low.

Leave a comment