Letter to Daniel Marlow on Exponent Error

On June 29, I sent a letter to Ted Wells, notifying him of the erroneous description of key figures in the Exponent report, but did not receive any acknowledgement.  In the presumption that Daniel Marlow of Princeton is more likely to be concerned about the erroneous research record (as well as having obligations that the research record be properly presented) I sent him a similar letter today, copying lawyers Daniel Goldberg and Jeffrey Kessler.

Dear Dr Marlow,

I am an experienced data analyst and have followed the Deflategate controversy with interest.  I wish to draw the following material misdescription and error in the Exponent report to your attention.  Will Happer, who I am copying, knows me and can vouch for me.  I am also copying this to Daniel Goldberg and Jeffrey Kessler.
In Figure 27, Exponent stated that the Patriot and Colt footballs were “set with the Logo Gauge” at 12.5 psig and 13 psig respectively.
However, this description is erroneous. The transients in the figure can only result from the footballs being set with the Master Gauge (NOT the Logo Gauge)  at 12.5 and 13.0 respectively.
The incorrect description is very material, because, if the footballs had been set with the Logo Gauge – as stated in the figure, the transients would be lower and the time interval of potential intersection with observations would be increased to approximately 6 minutes even at the 67 deg F initialization temperature assumed in the Figure (the temperature most adverse to the Patriots).  After correcting the error, it can no longer be said that anomalous Patriot deflation exists regardless of assumption on gauges.
While both you and Exponent have argued that the Non-Logo gauge must have been used for Patriot measurements, this does not justify Exponent’s erroneous description of the procedures used in Figure 27 (and 30) or any conclusions drawn from that figure. I have  described the issue in more detail https://climateaudit.org/2015/08/07/exponents-trick-to-exaggerate-the-decline/.
I find it very troubling that judgements have already been made based on figures that have been incorrectly described.   I draw this to your attention in the belief that you will also be concerned and will feel obliged to take steps to correct the research record as soon as possible. Obviously, the Deflategate  controversy has drawn very attention and is proceeding to a difficult decision, a decision that, in my opinion, is made more difficult by the misrepresentation of these figures in the Wells Report. In my opinion, there is some urgency that you correct the record expeditiously if you concur with my analysis, as I anticipate that you will.
I previously drew this error to the attention of Ted Wells on June 29 without any effect, but hope that you will have more concern.
Regards,
Stephen McIntyre
Climate Audit​
———- Forwarded message ———-

From: Stephen McIntyre
Date: Mon, Jun 29, 2015 at 9:28 PM
Subject: Exponent Report
To: Ted Wells at Paul Weiss

Dear Mr Wells,

While there have been many criticisms of Exponent’s technical analysis, I wish to draw your attention to a material and highly important misrepresentation of methodology in their technical report.  The calculation of transients, used in the important analyses of Figures 27 and 30, is described as follows:
In recognition of the remaining uncertainty as to which gauge was used to measure the footballs pre-game and in the interest of completeness, similar tests were run using the Logo Gauge. The Logo Gauge was used to set the pressure of two balls to 12.50 psig (representative of the Patriots) and two balls to 13.00 psig (representative of the Colts). From each set (corresponding to each team), one ball remained dry while exposed to the game temperature and the other was wet.
For reasons set out in more detail at Climate Audit (https://climateaudit.org/2015/06/29/exponents-transients-bodge-or-botch/), it is impossible that the bolded statement is a correct description of Exponent’s actual methodology. To arrive at the dry Patriot transient illustrated in Figure 27,  the Logo Gauge would have had to have been set to ~12.81 psig. It appears more likely that a different gauge was used to initialize the footballs in the simulations illustrated in these figures.   If Exponent had used the stated methodology, they would not arrive at the results illustrated in Figures 27 and 30.
The misrepresentation is material and warrants an immediate corrigendum,  together with retraction of technical analysis depending on this mistake. It is by no means the only statistical error in Exponent’s report, but is somewhat distinguished by it being an actual misrepresentation as opposed to a methodological critique.
I have considerable experience in statistical analysis.  I have made presentations to a panel of the U.S. National Academy of Sciences, a subcommittee of the U.S. House of Representatives and my work has been covered on the front page of the Wall St Journal, by CNN and Fox News and internationally. In 2010, the New Statesman magazine in the U.K. recognized me as one of “50 People Who Mattered” in 2010.
Regards,
Stephen McIntyre
Climate Audit

20 Comments

  1. Posted Aug 8, 2015 at 1:50 PM | Permalink

    Well there you go again Steve.

    It will be fun to see someone I know on the program when football minus Brady kicks off.

  2. michael hart
    Posted Aug 9, 2015 at 6:59 AM | Permalink

    Is there a word missing from the sentence
    “Obviously, the Deflategate controversy has drawn very attention…”?

  3. EdeF
    Posted Aug 9, 2015 at 11:17 PM | Permalink

    Hard to turn around a battleship at full steam ahead.

    • timg56
      Posted Aug 20, 2015 at 11:17 AM | Permalink

      Not difficult at all. Just give it rudder.

  4. Salamano
    Posted Aug 10, 2015 at 6:17 AM | Permalink

    That’s a weird-sounding accolade to end a letter with…especially if they don’t have any idea what it means. Your analysis ‘should’ matter though, since it seems like it hasn’t been done despite millions and millions on the line.

  5. MikeN
    Posted Aug 10, 2015 at 7:03 AM | Permalink

    This letter will not change anything.

    Perhaps he will mention a correction in a different forum, or even in the appendix of his next paper.

    Then we will be told the mistake doesn’t matter to the end result.

    The existing charts and the two minutes average will continue to be used.

  6. mpainter
    Posted Aug 11, 2015 at 10:04 AM | Permalink

    “In the presumption that Daniel Marlow of Princeton is more likely to be concerned about the erroneous research record (as well as having obligations that the research record be properly presented) I sent him a similar letter today, copying lawyers Daniel Goldberg and Jeffrey Kessler.”

    ###

    If your presumption is correct, your letter could cause a resolution of the issue. But, as I understand, Marlow is a consultant paid for his work by Paul, Weiss.

    I’m sure all would be interested in any reply by Marlow.

  7. John Faulstich
    Posted Aug 11, 2015 at 10:40 AM | Permalink

    The question is if Marlow ignores this, does Kessler quietly spread this around academia?

  8. Pat Frank
    Posted Aug 15, 2015 at 11:45 AM | Permalink

    It’s been a week and I gather no reply from Dr. Marlow, or we’d have heard about it. Guess he don’t got to show you no stinkin’ badges, Steve.

    • mpainter
      Posted Aug 15, 2015 at 1:12 PM | Permalink

      Daniel Marlow was contacted by the New York Times in the preparation of an article by James Glanz May 6, 2015, Marlow quoted:

      “Paul, Weiss has asked me not to comment and to refer questions to the NFL”

      Marlow will of course follow the instructions of his benefactors (he is their regularly paid consultant). I would imagine that his professional relation with Paul, Weiss nets him thousands. The world of money is watching. There is a good chance that he will be summoned as an expert witness.

      There are times when a scientist must think of himself. Can’t say that his behavior is not understandable.

      Apparently Marlow carries some weight among physicists in the world of academia, according to the article. Good read.

  9. timg56
    Posted Aug 20, 2015 at 11:19 AM | Permalink

    I just heard yesterday that the judge overseeing the case mentioned something to the effect the science evidence presented by the league was not necessarily that convincing.

    • mpainter
      Posted Aug 20, 2015 at 12:49 PM | Permalink

      I could find no such remark. Do you have a link, because if the judge used such an expression, it is the crack of doom for Roger Goodell.

      • kim
        Posted Aug 21, 2015 at 7:26 PM | Permalink

        The Boston Globe has a hearing transcript. It’s 83 pages too long for me to read.
        ========

        • kim
          Posted Aug 21, 2015 at 8:01 PM | Permalink

          Oh, well, skimmed it. Not much about the science, but ‘quantum leap’ on p. 6 is curious.
          ===========

        • mpainter
          Posted Aug 21, 2015 at 9:11 PM | Permalink

          I saw a legal blog where the 8/19 hearing was discussed. Judge Berman asked some more pointed questions of the NFL counsel and made even more pointed _remarks_ directed at same. It does not look good for Goodell.

          IMO, this is what Goodell is looking at, possibly: Goodell is pushed aside and Bradey’s suspension is suspended while an arbitrator appointed by the judge re-opens the whole issue, allowing all rights to Bradley and the NFLPA to examine documents and take testimony from whomever, including Goodell. Goodell will come out with a foul odor clinging to him and Bradey will have all he needs to proceed with a successful defamation suit against Goodell. Everything depends on how dumb Goodell is, and he might just be dumb enough to risk such an outcome.

  10. John Faulstich
    Posted Aug 21, 2015 at 8:44 PM | Permalink

    so do we think the (to the eye) small errors in the transients were intentional or mistakes? http://www.robertblecker.com/deflategate-the-smoking-gun/

  11. John Faulstich
    Posted Aug 21, 2015 at 9:21 PM | Permalink

    I think Goodell weighed his options and decided that he could not afford to back down as questions would arise as to why, so he thinks he either wins or loses in court, and if he loses it’s based on the process which will hurt him some, but not on the facts, which would hurt him a lot. But there are some saying since the League went to the court requesting it to basically validate the decision, they risk the judge ruling on the facts as well as the process. I think brady has ammo for the defamation lawsuit and I hope he uses it.

    • mpainter
      Posted Aug 23, 2015 at 4:57 PM | Permalink

      I read the 8/19 hearing transcript, all 83 pages, skimming over some. Kessler argued law and precedent, and the bench seemed impressed. Nash, NFL attorney, basically argued that Goodell acted within his authority and discretion. The bench took issue with some of it, obviously not much impressed by NFL arguments.

      It seems obvious that the NFL is in a very weak position for settlement talks, the bench’s attitude having been so obvious. IMO, NFLPA/Brady will not feel the need to cede anything to the NFL. They can hope for the best from the court. It seems that the bench is doing Goodell the favor of extracting himself from the mess he has created.

      • mpainter
        Posted Aug 23, 2015 at 4:59 PM | Permalink

        Rather, doing Goodell the favor of providing the opportunity to extract himself.

  12. Ian
    Posted Sep 10, 2015 at 2:15 PM | Permalink

    A bit late for this thread, but there’s a parallel story on Formula 1 tyre pressures:

    An unexpected alarm

    That margin was again clear (gap to next driver) when Hamilton’s engineer Peter Bonnington came on the radio with seven laps to go and told him he needed to put in some flat-out laps – or “hammer time”, as he put it.

    Hamilton was flummoxed by the request, but Bonnington merely replied: “Don’t ask questions; just execute.”

    The need for this became apparent just before the end of the race, when it emerged that Mercedes had been referred to the stewards because the starting pressures of the left-rear tyres on both cars were below the minimum permitted by Pirelli on safety grounds.

    In Hamilton’s case, the margin was only 0.3psi, a 1.5% margin that engineers from rival teams admitted would make almost no difference, while adding that that was hardly the point. Rules are rules, they said. You’re either within them or you’re illegal.

    Except in F1 it is not always that way.

    In the past few years, governing body the FIA’s approach to finding cars have features that break the rules has tended – but not always – to be to warn the team in question and tell them to change the part. Either that weekend or – in some cases – for the next one.

    This, though, was a different scenario; it was ruled that although the tyres were technically not in compliance, Mercedes had not done anything wrong.

    The investigation by the stewards revealed not only that the tyres had been above the minimum pressure when fitted to the car but that the tyre blankets that maintain the temperature – to which pressure is directly related – were unplugged, “significantly below” the maximum permitted temperature, and “significantly different” from those of other cars.

    On these grounds, the stewards declared that Mercedes had “followed the currently specified procedure for the safe operation of the tyres”.

    There were cynics who dismiss the verdict as a fudge, but Mercedes F1 boss Toto Wolff said he could “absolutely rule out” any deliberate foul play, such as setting the temperatures at a point that met the minimum when they were measured but allowed them to drop below that point for a competitive advantage by the time the race had started.
    Play media
    Jump media player
    Media player help
    Out of media player. Press enter to return or tab to continue.

    Rosberg’s world title hopes up in flames?

    “We have worked the whole week with Pirelli to find solutions to make the tyres safe, and we were trying to guide them on minimum tyre pressures and minimum camber,” said Wolff.

    “I can rule out that we would be the ones who would try to gain an advantage in a way that is absolutely unscientific and uncontrollable.

    “How would you want to measure how much the tyre pressure drops once you disconnect the blanket? Why would you have it only one on tyre and discrepancies between the tyres?”

One Trackback

  1. […] using the Logo gauge (Figure 27.).  A significant error in that graph setting is described in a report in another blog by Steve McIntyre.  Fixing that mistake may result in a similar effect as to the gauge switch […]

Follow

Get every new post delivered to your Inbox.

Join 3,586 other followers

%d bloggers like this: