Last week, we solicited your questions for Tobias J. Moskowitz and L. Jon Wertheim, the authors of Scorecasting: The Hidden Influences Behind How Sports Are Played and Games Are Won.
You shot them a lot of good questions — heavy on the NFL, to be sure. (Related: we’ll be hosting a Q&A here tomorrow in all likelihood with George Atallah of the NFL players’ union: so get your lockout questions ready.) Their responses are below. Thanks to Toby and John, and all who participated.
First, thanks to all of you for responding and posing some very interesting (and some very tough) questions. We wrote the book for people like you, who enjoy exploring the unconventional and who think about things a little differently. While we didn’t have time to answer all the questions – nor do we have all the answers – our aim is to foster debate and discussion on these and other subjects in sports. As we said in the introduction to our book, “we hope to settle some bar fights, but with any luck we’ll start a few more.” So, here goes.
Jimmy Johnson created (at least he claims) a “value chart” for the NFL draft. Is there any independent investigation of these numbers and, if so, how valid are they in terms of long term player productivity, etc.? —Craig A.
In fact, we devote a chapter of our book to this. Let the record reflect: it was Mike McCoy – a long-time partner of Jerry Jones dating back to their days as oil speculators – who created “the chart.” Jimmy Johnson, as Jones’s first hire, used the chart to great effectiveness in exploiting other teams on Draft Day. “The chart,” as everyone around the NFL came to call it, helped identify which teams overpaid for draft picks consistently, and hence which teams you’d want to trade with most often. As the Cowboys under Jones and Johnson became more successful, “the chart” became something of a sacred text. When Dallas assistants were hired away, they brought copies of The Chart with them and eventually everyone was following the same valuations.
However, the chart, which was a crude calculation made by McCoy at the behest of Jones, was never checked. When you look at the data, the chart consistently overvalues top draft picks. That is, a top pick at a position might get 50 to 100 percent more (in terms of salary or other players) than the next pick, but be only slightly, if at all, better in terms of his career performance numbers. This is because the chart McCoy calculated was based on what teams did and not on what teams should do. We run through the numbers in the book and highlight some poignant examples. There are now several teams going “off the chart” and reaping benefits from doing so — including the Cowboys, the team that created it in the first place.
Do you like the NFL’s new post-season overtime setup? Do you feel that it’s significantly more equitable than the traditional “sudden death?” And how do you feel about this new NFL plan vs. college football’s OT rules? —Cubnut
We don’t like the new (and almost risibly convoluted) NFL post-season OT setup and feel it’s only slightly better than the previous sudden-death format that they still use during the regular season. The college football OT rules are better. But our favorite suggestion came from the Quanbeck brothers. They suggested auctioning off field position versus receiving the ball. In other words, if you want the ball first, I’ll give it to you if you take it at the 15-yard line. Two teams could approach midfield with sealed bids of field position of where they’ll take the ball first. The lowest one gets it first, and the two teams play sudden death from there. Just imagine the fanfare you could add.
Given the huge signing bonuses required to secure the top few picks of the NFL draft, getting these picks wrong can damage a team quite badly (e.g., JaMarcus Russell). My two questions are:
1) How do draft positions rank in terms of expected value for the team?
2) If you were designing the Rookie Salary Scale that is possibly/probably coming in the new NFL CBA [collective-bargaining agreement], what would it look like?
We assume you’re talking only about the NFL. To answer the first question: not that well. It’s very hard to figure out whether you’re getting JaMarcus Russell or Sam Bradford with the top pick. Here’s what you do know for sure: you will pay through the nose for both of them. So, what we find is that teams are better off trading away top picks for more lower picks. The top pick is paid the equivalent of three or four picks in the late 1st to early 2nd round. Yet, having three or four late first-rounders is far more valuable on the field than a top pick. The chance that at least one of them becomes an All-Pro is higher. Plus, as you mention, if you draft the No.1 pick, owners and fans expect to keep investing in him, which doesn’t give his backup much of a chance. This compounds the mistake. A lower-drafted player isn’t afforded as many chances and isn’t overinvested in when he is clearly a bust (e.g., Russell).
On the myth of the hot hand….my question involves “the zone” where an athlete experiences a time frame of superior performance. If you believe in the zone (I do), doesn’t that verify that the hot hand exists? —frankenduf
Anyone who has played sports – even at the most recreational level – feels as though they can “feel it” on some days – and some days you don’t. What we found, however, is that most streaks in sports are consistent with random chance. Recent past performance doesn’t seem to be predictive of future performance. That doesn’t means “zones” couldn’t exist. It simply means that it’s hard to find evidence that is inconsistent with random chance. In the book, we use an example of .333 hitter. It doesn’t mean that he’ll get a hit once out of every three at-bats. It simply means that, over the course of a season, he’s getting a hit 33.3 percent of the time. In those 600 or so at-bats, he’ll go 0-for-10, and he’ll go 5-for-5. Does that means he’s in and out of the zone? Or is this simply what you’d expect? Again, we’re just not convinced it’s much different from random chance.
Watching a football game dominated by one team, I frequently hear the announcers in the second half state something like “the (losing team) defense is worn out after being on the field so long.” Why doesn’t the (winning team) offense get tired? —Zifmia
Great question. It’s a bit like “defense wins championships,” a bromide we address in the book. Defense isn’t glamorous or glorious, so we assume it creates fatigue in a way that offense doesn’t. Your point, though, is well-taken. For the sake of argument, you could submit that defense is more physically taxing. But we say that, if anything, it’s more mentally taxing. You’re not seeking a gain, so much as you’re aiming to protect the status quo.
Should basketball teams sub out players in foul trouble? Absent the player needing an emotional break from the game, doesn’t subbing a guy out voluntarily accept the penalties for fouling out rather than only suffering them if the risk is realized? While some may argue that late game minutes are more important than other minutes, there is no evidence of this: points are points are points. Build a big enough lead, and the end game is meaningless. Why willingly take a player out of the game for ten minutes in the same half in hopes that you won’t lose him for five minutes at the end of the game? —BSK
We actually analyze this in the book, and you’re exactly right. The premise that coaches should sit a foul-plagued player until late in the game is faulty. We crunched the numbers and found that coaches sit star players in foul trouble for way too long. We also found that – because of official bias – star players are even less likely to pick up the sixth foul.
Will Nancy Lieberman be the first woman NBA coach? —Ames
Love the nickname “Lady Magic.” But our money is on Rebecca Lobo, not least because she married a sportswriter.
We constantly hear that a given player’s stock has risen/fallen as a result of performance at the NFL Combine (faster 40, higher vertical leap, etc.), but has anyone actually been able to tie combine performance to on-field outcomes? Are any of the tests at the combine more predictive than others in determining the value of a player? —Chris W
Funny you mention this. We’re looking at this (for the sequel), and Moskowitz is writing a paper on this with Dick Thaler. Without revealing too much, it turns out that teams put more weight on these combine figures than they should. Combine performance doesn’t seem to be commensurate with future performance. What troubles us is the exercises themselves. We question how running a straight sprint in no pads (with no defender) is predictive of a receiver’s in-game performance. Stay tuned!
How long will Tim Lincecum‘s hair be next season? —Andrew
Within two standard deviations of Brian Wilson‘s beard length.