Which is why I want to know why so many people think it's outlandish that the actual facts under discussion (Jeter's plays made) could be due primarily to Jeter's ability to play shortstop.

Because you haven't actually demonstrated that it is the case. All you have done is noted that he gets to less balls than the average SS and decided that a set of reasonable explanations that has generally been accepted doesn't suit you. Providing some factual basis instead of inference that these balls are actually being hit to SS, and that they are actually fieldable, and that Jeter is actually missing 2 real balls every five games instead of 2 theoretical balls every five games, and people will be a lot less skeptical.

As for (b), Tom Tango does his WOWY analysis, where he compares Jeter to the Yankees backup shortstops, and comes to similar conclusions as Guy has.

What's that add up to? 5-10 games a season maybe? That's not very convincing data.

But who do you mean by 'everyone'? Jeter's bat makes up for his defensive miscues, so the Yankees were never going to bench him (rightly so).

If Guy's assessment accurately reflects reality, has Jeter really been helping his teams at SS, even given his excellent offense for the position? If so, how much has he been worth? How much should we chop off of his career WAR, for example.

We've had scores of young players who were drafted as SS, came up through the minors as SS, and once in the majors, were shifted to 2B or 3B, because they couldn't hack it at short. Since plenty of these players turned out to be good glove men, wouldn't it follow logically that they were shifted due to lack of range? If this is the case, how much worse would these guys be than Jeter? 5 runs? 10? 25?

But how many of those players were DEREK JETER winner of championships and leader of men? Seriously, Jeter is a unique case. He very quickly cemented the job of Yankees SS and even when a clearly better SS was added to team Jeter remained the SS.

To answer your question, I bet a fair amount of those players you describe were or are better defensively than Jeter, they just aren't Jeter.

So, who exactly has been so wrong about Jeter?

I remember several years ago when the "Hmmm, Jeter may suck at defense" stories were starting to break, at least in the Rob Neyer circle, I remember reading comments from scouts that Jeter was a good defender.

Now, maybe that was isolated and most people did think Jeter wasn't good, but just didn't want to be quoted on it or say anything about it. But I think some teams did and still do have a long way to go as far as analyzing defense goes.

Is that worse than a $25 million 3B who plays 120 games a year? I think the real issue is his health, not his fielding.

Not sure where you're getting 120 games from. He has played that few number of games precisely never, and has come close to it just once (124 games) in his tenure with the Yankees. The fact that he plays 7 or 9 or 12 games a year at DH doesn't appreciably change anything.

If you're arguing that he will only be able to play 120 games a year going forward if he stays at third, I don't see what you're basing that on.

Because you haven't actually demonstrated that it is the case. All you have done is noted that he gets to less balls than the average SS and decided that a set of reasonable explanations that has generally been accepted doesn't suit you

What is the reasonable explanation? I think the problem that Guy/CW have mentioned is extremely likely - basically, that BIS/UZR type data is very likely corrupted by the fact that whether or not they think a ball is in a fielder's zone has to do with how good of an attempt a fielder made. Jeter is poor up the middle, and "past a diving Jeter" was a catch phrase on this site for a reason. Those are the kind of plays he struggles with. I also think that balls he gets to in the hole, that some shortstops just plant and gun while he does the jump throw - I think those are likely to be scored incorrectly by a BIS player, because you don't have Elvis Andrus ghosting the play, planting, and firing a bullet to first at the same time to see if the jump throw gave the runner an extra half step.

Because you haven't actually demonstrated that it is the case. All you have done is noted that he gets to less balls than the average SS and decided that a set of reasonable explanations that has generally been accepted doesn't suit you. Providing some factual basis instead of inference that these balls are actually being hit to SS, and that they are actually fieldable, and that Jeter is actually missing 2 real balls every five games instead of 2 theoretical balls every five games, and people will be a lot less skeptical.

Those explanations are being supported by subjectively collected data that is influenced by a number of scorer biases - if the data is wrong (and I have shown why I think that it is), the explanations lose a lot of their support and what we are left with are the facts. And the facts are that Jeter doesn't make that many plays.

It's worth noting that it's hardly just Jeter on which the facts disagree with UZR and TotalZone. Even if you throw out 16,000 BIP worth of data on Jeter, you have to explain why the same effect of range bias in the "advanced" fielding location data is seen on the rest of the player population, too.

Tom Tango does his WOWY analysis, where he compares Jeter to the Yankees backup shortstops, and comes to similar conclusions as Guy has.
What's that add up to? 5-10 games a season maybe? That's not very convincing data.

Actually, WOWY compares Jeter to all other SSs who have played behind his pitchers. So, for example, it compares how many outs Jeter made behind Roger Clemens to the outs made by other SSs who played behind Clemens (in NY, TOR, BOS, and HOU). So it's not based on Jeter's backup SSs, which would indeed be a small sample. WOWY also adjusts for opposing hitters, and makes an age adjustment for pitchers (in case their GB/FB distribution changed). Basically, it adjusts for everything that could materially affect the BIP distribution.

Providing some factual basis instead of inference that these balls are actually being hit to SS, and that they are actually fieldable,

If you assume CW has good-faith reasons to doubt the accuracy of the BIS data, how exactly could he do that? Your argument ends up being perfectly circular: you won't believe any evidence except the evidence you I know supports your preferred conclusion.

So if we can agree to set aside the BIS data (not asking you to concede it's wrong, just consider other evidence), what do you think could explain such an odd distribution? Why would Jeter get so few opportunities, while Yankee 2B and 3B aren't similarly impacted? And are there any other players who have seen such an unusual distribution over a long career? That would certainly bolster Jeter's case.

What is the reasonable explanation?

That Jeter doesn't get as many opportunities due to staff construction, the Yankee way of pitching (which has been fairly consistent since Micheals took over, even if the coaches and personal have changed over the course of time), and the third baseman poaching some of his balls.

I think the problem that Guy/CW have mentioned is extremely likely - basically, that BIS/UZR type data is very likely corrupted by the fact that whether or not they think a ball is in a fielder's zone has to do with how good of an attempt a fielder made.

Corrupted by 10 to 30 runs every year? That seems unlikely. In fact, based on reading the Primer and RLYW game chatters, I would guess that when people already "know" a player is a bad fielder, they are more likely to assume most balls he doesn't get should have been fielded.

Those explanations are being supported by subjectively collected data that is influenced by a number of scorer biases - if the data is wrong (and I have shown why I think that it is)

Yes, you've shown how you convinced yourself that it's true, by making inferences on human perception and camera angles. But you haven't shown it to be inaccurate, you've just inferred that it must be because of some guesses based on generalities of human perception. I remain unconvinced.

And the facts are that Jeter doesn't make that many plays.

But these are facts without context, which makes impossible to know what they really mean.

Why don't you guys just come straight to the source and ask me if I could field more balls?

Not sure where you're getting 120 games from. He has played that few number of games precisely never, and has come close to it just once (124 games) in his tenure with the Yankees. The fact that he plays 7 or 9 or 12 games a year at DH doesn't appreciably change anything.

If you're arguing that he will only be able to play 120 games a year going forward if he stays at third, I don't see what you're basing that on.

Because he's 35 years old? He's probably not about to start getting healthier. He's averaged 133 games the last three years, but only 122 starts at 3B, and you've got to expect that to continue to go down as he gets older. He hasn't topped 600 PAs in a season since 2007--I know the idea of a $25 million DH sounds unappealing, but at some point the Yankees are going to have to bite the bullet if they want to protect their investment.

Actually, WOWY compares Jeter to all other SSs who have played behind his pitchers. . So, for example, it compares how many outs Jeter made behind Roger Clemens to the outs made by other SSs who played behind Clemens (in NY, TOR, BOS, and HOU).

And what good does that do? Roger Clemens was a very different pitcher in 1986 than he was in 1999 than he was in 2007. His peripherals weren't the same, his repertoire wasn't the same, his approach to pitching wasn't the same.

WOWY also adjusts for opposing hitters, and makes an age adjustment for pitchers (in case their GB/FB distribution changed). Basically, it adjusts for everything that could materially affect the BIP distribution.

Any time a metric makes a general adjustment for the entire population of players I am extremely skeptical when it is applied to a single player, or a single set of player. Just like FIP, or the aging curve or anything else that claims to qualify the performance of everyone, its going to fall apart when applied to many individual players.

If you assume CW has good-faith reasons to doubt the accuracy of the BIS data, how exactly could he do that?

It's not my burden. If you want to claim that all the data we have is faulty, it really should be up to you to actually show it's faulty.

Your argument ends up being perfectly circular: you won't believe any evidence except the evidence you I know supports your preferred conclusion.

Yours is equally circular: Jeter makes less plays and we know that he has the same opportunities because all the evidence that shows he has less opportunities must be flawed.

And neither represents my preferred conclusion. I would prefer Jeter was a good defensive SS. But that isn't the case and I've accepted that because there is a ton of objective data that shows he is on average about 10 runs worse than average year. What I won't do is accept a system that is way out of line with what is already out there and offers absolutely nothing to support the contention that the original data that has been in use for more than a decade is corrupt. If you have irrefutable objective data that shows Jeter is 20 runs worse than average a year, I'll accept it. As of right now, there are just way too many flaws in your method to convince me and plenty of other people who aren't big fans of Jeter.

So if we can agree to set aside the BIS data (not asking you to concede it's wrong, just consider other evidence), what do you think could explain such an odd distribution?

Why? We've done this at least half a dozen times over the last year or two. You don't agree with what I think about this and I don't agree with you.

#116, I suppose you can make a case that his recent hip troubles combined with some deterioration in his fielding and batting numbers, along with more missed games, could point to making him a full time DH.

I agree that's reasonable; I still would prefer not shunting him to DH until it becomes clear his body can no longer hold up playing the field. I don't think we're close to being there yet.

It's not my burden. If you want to claim that all the data we have is faulty, it really should be up to you to actually show it's faulty.

Go back to the previous page. Read all those links, where I go into detail as to the flaws in the data caused by scorer bias. How haven't I shown there are faults in the data?

I have very little to add to this discussion. But I would like to thank those participating for having it, because it is very interesting.

FWIW (which is nothing), GuyM's and CW's theories make a lot of sense to me, probably because they support my anecdotal view of Jeter's defense developed over watching him in hundreds of games. I have no idea if it is actually correct, but the concept that he is not being charged with missed opportunities because they do not actually look like opportunities (due to range, visible effort, whatever) seems like a very reasonable hypothesis. And the cited data seems to at least lend some support to the hypothesis.

Popup: you don't appear interested in considering the evidence or making a good faith argument. You say "all the evidence" that Jeter had fewer opportunities, but the only evidence of that is the BIS data Colin is discussing. There is no other direct evidence, except the STATS data. And when UZR used that data, it rated Jeter very poorly, similar to what WOWY shows.

You are "skeptical" of Tango's WOWY article, but have you even read it? (click on article on page 44). It sure doesn't sound like it. Nothing you say raises serious questions about his methods.

You seem to feel there is some widespread consensus of analytic methods that Jeter is -10, while suggesting -20 is an outlandish and crazy notion. But there is really only one piece of evidence for the -10, and at least equally good evidence for -20. You choose to believe the BIS data, although you don't seem to know much about it and you have no remotely credible theory to explain it's conclusions. But please stop acting as though yours is the only plausible view, or that everyone else has some burden of proof that you don't have to meet.

ou don't appear interested in considering the evidence or making a good faith argument.

Yeah, you're probably right. The reading for this discussion has already bored me to death and eaten a huge chunk of time I needed for work. I'll bow out, I just don't care enough or know enough about the subject to participate in the conversation.

Fair enough. But when you have the time I do suggest you read Tango's chapter on WOWY -- it's a quick read and a really interesting/important study.

If Guy's assessment accurately reflects reality, has Jeter really been helping his teams at SS, even given his excellent offense for the position? If so, how much has he been worth? How much should we chop off of his career WAR, for example.

The difference between Rally's WAR estimate (based on Total Zone) and Tango's WOWY analysis would be in the ballpark of 15 runs per season, or a loss of about 22 WAR from Jeter's current total of 70.

The difference between Rally's WAR estimate (based on Total Zone) and Tango's WOWY analysis would be in the ballpark of 15 runs per season, or a loss of about 22 WAR from Jeter's current total of 70.

Which would mean that he's not a HOFer. I don't think anyone really believes that.

Which would mean that he's not a HOFer. I don't think anyone really believes that.

Holy hell in a handcart, is Ray arguing from conventional wisdom on who is and isn't a Hall of Famer? Now I've seen everything.

Which would mean that he's not a HOFer. I don't think anyone really believes that.

Thanks, that's priceless. A perfect logical falacy: "Hmmm, X is an interesting idea, which seems to have some solid evidence behind it. But X would mean Y, and nobody believes Y. So X must not be true."

And FWIW, I don't agree that accepting Jeter's real defensive shortcomings necessarily means Jeter isn't HOF material. It's certainly possible that Jeter has been a bit better than his WOWY estimate, so maybe he's a 60 WAR player. One could argue that it's the Yankees' fault he was kept at SS all those years, and we shouldn't hold him 100% responsible for the consequences. And if you give him credit for his incredible post-season record, he would still have a very good HOF case.

But it's certainly true that Jeter is much closer to being Willie Randolph than Mickey Mantle, in terms of career value.

Tango's article linked in #121 has 4 separate estimates of shortstop defense. Depending on which one you use he's below average by (per season):

pitchers - 38 plays/29 runs
batters - 25 plays/19 runs
ballparks - 18 plays/14 runs
baserunners - 11 plays/ 8 runs

Only the pitcher one is inconsistent with John Dewan's rating (-14 runs per year from 2003 on), though TZ and UZR rate Jeter better than that. All of the wowy ratings look better than using his raw assist totals if that puts him at -60 plays per year. So Jeter may really be a magic pixie that repels ground balls.

Thanks, that's priceless. A perfect logical falacy: "Hmmm, X is an interesting idea, which seems to have some solid evidence behind it. But X would mean Y, and nobody believes Y. So X must not be true."

What you're doing with his defense is drawing every possible inference against him. I'm not a big fan of that kind of analysis, and I don't accept it as valid.

Defense is not something to express near certainty over to the degree you're doing.

Yeah, you're probably right. The reading for this discussion has already bored me to death and eaten a huge chunk of time I needed for work. I'll bow out, I just don't care enough or know enough about the subject to participate in the conversation.

I completely didn't expect this response, given that this is the internet and all. You are to be commended.

Holy hell in a handcart, is Ray arguing from conventional wisdom on who is and isn't a Hall of Famer? Now I've seen everything.

Thanks, that's priceless. A perfect logical falacy: "Hmmm, X is an interesting idea, which seems to have some solid evidence behind it. But X would mean Y, and nobody believes Y. So X must not be true."

Ray didn't say "so X must not be true", nor did he argue anything. He stated that what you have appears to go against conventional wisdom. Although he didn't say it, one could infer the suggestion that it's worth considering other possibilities. I infer the same from your knee-jerk defensiveness to anything remotely resembling criticism.

EDIT: I see that Ray has stated what I've inferred. And that it's not far off from what MCA has been saying on the prior page.

EDIT2: Wow, I just saw the time gap between #128 and my #129. I really need to refresh the page before posting something I wrote before going into a meeting.

All of the wowy ratings look better than using his raw assist totals if that puts him at -60 plays per year. So Jeter may really be a magic pixie that repels ground balls.

AROM: if you've read this thread, you know I've said multiple times that I don't believe the raw stats tell the full story. I just think they're an important starting point.

Tango later updated his WOWY analysis, and has Jeter at about -33 plays (-23 runs) per season through 2008. That combines the various models, mainly pitchers and hitters (for various reasons the ballpark model is not really correct -- Jeter ends up being compared to himself too much, IIRC). Using raw stats, Jeter is about 50 outs below average. So WOWY is consistent with Jeter having somewhat fewer opportunities than usual. I can believe that. (Of course, it could also be that Jeter takes more than his share of pop-ups, in which case WOWY is overstating his fielding performance.) What's very hard to believe is that Jeter has been only -7 runs in range, the claim of TZ, while making 50 fewer plays than average

Look, if you want to say Jeter has been -14, then we're down to a 9-run disagreement. Still significant, but getting closer. And that means Jeter loses 8 WAR from his current total -- 10 if Dewan isn't counting double plays (don't tell Ray!).

What you're doing with his defense is drawing every possible inference against him. I'm not a big fan of that kind of analysis, and I don't accept it as valid.

That's not true at all. The worst interpretation of Jeter is to just use his outs total and leave it at that. I've gone as far as suggesting that half of Jeter's shortfall might be explainable in terms of something other than poor fielding (although that's not very likely).

And I certainly haven't claimed "certainty." It's Jeter's defenders who seem certain that UZR is correct, and only the BIS-based version of UZR at that (before MGL changed data sources, Jeter's UZR ratings were much worse.)

Ray didn't say "so X must not be true"

When Ray said "I don't think anyone really believes that," I did infer he meant that it was not true, rather than simply making an observation about the perception of all people. If I'm wrong, I happily retract my comment. But I would note that he didn't take the opportunity to make that claim.

And FWIW, I don't agree that accepting Jeter's real defensive shortcomings necessarily means Jeter isn't HOF material. It's certainly possible that Jeter has been a bit better than his WOWY estimate, so maybe he's a 60 WAR player.

I think we've gone over this about twenty times already, and I don't think we're going to convince each other. But once more with feeling...

Jeter's WOWY defense estimate is -24 runs per season. Extrapolating from the 60 WAR estimate, you suggest that Jeter's true defensive value could be more like -17 runs per season - his Magical Pixie Power could account for seven runs per season. My question is, why do you accept that ball distribution and positioning could account for seven runs per season, but you're nearly certain ball distribution and positioning couldn't account for fourteen runs per season?

My position is that we simply do not know what a reasonable estimate is of the effects of ball distribution and positioning on plays made. Even when we do all sorts of (very cool) work like the WOWY system to account for various causes of differences in ball distribution, we still don't know what level of confidence we should have in a single player's results. You suggest that error bars on the range of seven runs per season are reasonable, even over a 15-year career. I don't see what the evidence is for this claim about reasonable error bars - it seems like one could just as easily claim the error bars should be set at three runs per season or fifteen runs per season. There's no evidence to back up any of those three claims. As such, I persist in my agnosticism.

EDIT: to clarify on WOWY, I think it's a useful system. I think we know more about defense because Tango did that work. However, the fact that a system is interesting and useful does not tell us that its results for any single player must be correct within a certain error range. Given how little it seems we really know about defensive statistics, a system could be very cool and very useful and still have results for individual players which are off from their real value by quite a wide margin.

Let's back the truck up a second, MCoA. You said:

Your theory requires play-by-play data to be quite massively compromised.

And, independently of Jeter, I believe this to be utterly correct - the batted ball data is compromised, in more ways than I think we've been able to enumerate and in no way that we've been able to properly account for. So any analysis using the data is suspect. And due to the proprietary nature of much of the data and the collection process, I don't know how much further we'll be able to move the ball in that direction.

So the question boils down to - how do you evaluate ANY player's defense, if you can't trust that data? Once you answer that question, you're on the right footing to evaluate what you think of Derek Jeter.

I expect Jeter has at least one more 300/370/440 type line in him,

I agree with this, too, except that this will be the year he gets hurt and plays 110 games. Otherwise, no.

CW-

After reading your articles, I'm very skeptical of the current pbp data. You showed, among other things, that BIS and MLBAM are getting very different total numbers of ground balls and fly balls and line drives, year to year. That's very bad, and suggests we need to be very careful when extrapolating from that data, if we're going to do so at all.

What I don't get, though, is why we should be any less careful extrapolating from "plays made" data. We already know that a million things other than player skill or player value are captured in "plays made". (WOWY demonstrates that, for one thing.) Given that, I don't see any reason to presume that we have the ability to turn plays made data into player skill / player value metrics that will nearly always measure player skill / player value within a certain range. That's the thing I've been arguing about.

So the question boils down to - how do you evaluate ANY player's defense, if you can't trust that data? Once you answer that question, you're on the right footing to evaluate what you think of Derek Jeter.

I think, given how problematic our data is, we need to consider another very problematic data set - the opinions of a player's contemporary observers. We should be very careful drawing any significant conclusions from the contemporary-opinions data set - as I said, it's certainly deeply flawed too - but I'd want to consider that data just as much as the various statistical measures.

My starting point, then, would be that when a player has remained a shortstop for 15 years and won several gold gloves, we should have a null hypothesis that he's pretty good. When we have various other problematic data suggesting he's either terrible or not very good, we're just ####### stuck and we don't know anything. If pressed, I'd probably give the most weight to contemporary opinion, and then some to the different defensive measures, and I guess I'd end up saying Jeter's maybe average or somewhat below average. Most importantly, I'd say I've got no clue how good Jeter is as a defensive player.

Jeter, I should note, is a really hard case. Most players who appear to be good defenders, who are treated by their teams as good defenders, also have good statistics, and vice versa. With those players we can be much more confident that we have a sense of how good they are, defensively. Jeter just happens to be a case where I think we should throw our hands in the air and admit we're still mostly ignorant.

EDIT: softened argument in paragraph three.

So the question boils down to - how do you evaluate ANY player's defense, if you can't trust that data?

I look at what the available defensive systems say and try to in effect give the player a grade within a couple of letters (e.g., A-B) without being too serious about that evaluation. (EDIT: After reading Matt's #135 I agree that it doesn't hurt to also consider available scouting reports and contemporary observation and the like. We're really suffering from a lot of darkness here.)

It's why when looking at specific players I like to start with VORP and then mentally adjust for defense. The problem with simply relying on something like WAR is that it locks you into one system. (And as I've stated on multiple occasions I particularly have reservations about WAR's defensive numbers for players before the advent of PBP systems, and I think calling different things for Yount and Jeter both "WAR" is highly problematic.)

I think we have to accept we can't take any defensive numbers too seriously, even now. And this entire discussion shows that.

CW and GuyM-

One final note. It seems to me that, rhetorically, you're using the evidence that the pbp data is compromised to suggest that the pbp data is compromised in the particular way that GuyM suggests it's compromised - that is, pbp data overrates bad defenders and underrates good ones. In none of the articles that CW linked - at least those I had access to read - was there any evidence that GuyM's hypothesis is correct, and that as such we should presume that pbp data systematically overrates Derek Jeter.

Without any evidence, GuyM's hypothesis isn't particularly convincing to me. I have no idea whether scorers will have that bias - my memory is that people like Chris Dial, who have actually scored tons of games, think that scorers don't have such a bias. So I take the pbp data as another problematic data set, but I do not take it as a data set that is probably systematically overrating Jeter.

Well, you can triangulate.

One leg of the triangle is the overall evidence that there is scorer bias that isn't being caught.

One leg is the overall spread of fielding talent observed by systems using batted ball data, compared to what we see in systems that do not incorporate such bias, like WOWY. (Not only WOWY, mind you, but that's in the conversation.)

And one leg is the correlation we see between a player's expected chances and his rate of converting those chances. (As well as a persistence in a player's rate of expected chances, even when he switches teams.) In other words, what we see is that a player's chances to field a batted ball, as measured by this sort of batted ball data, seem to persist over a period of time in a way that's attributable to the indentity of the player, not his environment. This isn't something I've published formally yet, but I do take a cursory look at it here.

Range bias seems to be the most plausible explanation for that last fact, although I am willing to believe that reputation bias is another potential factor.

Most importantly, I'd say I've got no clue how good Jeter is as a defensive player.

I don't think saying "defensive statistics are suspect, we don't really know the truth, so let's assume Jeter is average or better" is a fair summary of our knowledge. I think you are underestimating two important factors. One is the value of knowing many outs fielders have made. There is no margin of error there -- we know the exact answer. That's the job of fielders -- to make outs. And Jeter does that less than almost any other SS in history. Now, I'm all for efforts to try to determine how much of that shortfall can be explained by factors other than his skill. But you would be MUCH closer to the truth just relying on outs made than assuming, against overwhelming evidence, that Jeter is average.

Second is the value of a large sample size. With Jeter we have 60,000 BIP, given up by dozens of pitchers and hundreds of hitters. That greatly reduces the chance of Jeter having a radically different number of opportunities than other SS. Look at his BRef page and see how many things are exactly league average for Jeter: RHB%, GBIP%. In any year, that may not be true. But over 15 years, these things tend to average out. And we DO have some idea how much the distribution of opportunities can vary for a player. The WOWY analysis calculates how different a fielder's group of pitchers can be. No one with a long career ends up with opportunities as skewed as Jeter's would need to be for him to be average.

Saying Jeter is average means saying "We should assume that Yankee pitchers have given up a bizarely maldistributed balls in play. Even though we don't have very good evidence for that conclusion. Even though there is no plausible theory to explain that outcome. Even though Yankee 2B and 3B don't have the same experience. Even though we know of no other fielder who has had this experience. Despite all that, it is a more reasonable assumption than believing Jeter has been a very weak fielder." And that's just not a reasonable assumption, given the evidence we have available.

You can reject my hypothesis for why the PBP data is wrong. It's just a hypothesis. But there really shouldn't be any question at this point that it is wrong.

GuyM - can you please answer my italicized question in #132?

You accept that the "bizarre maldistribution" could account for more than a hundred runs over fifteen years, even without a "plausible theory" to explain why. (I've offered a couple theories I find perfectly plausible, for what that's worth.) Why do you think the WOWY data could be off by a hundred runs, but not by two hundred or three hundred?