Open letter to Robert Kraft. Delivery receipt requested.

( Link to home page:  leading in particular the material that Robert Blecker referenced in his wbur op-en piece)

As far as I know, Robert Kraft has not seen this.  Any help getting it delivered and confirming that he read it would be greatly appreciated!

Letter as of 8/31/2015.

Robert Kraft, owner New England Patriot

Re: How to exonerate Patriots and restore draft picks

Mr. Kraft,

Thank you for making the Patriots a success.

When you accepted the punishment, I believe you recognized that you were in an unwinnable battle of perceptions with the other owners, so you elected to face that reality and move on.  Perhaps you were keeping your powder dry for round two when cooler heads could prevail.  This letter should help with that.

The League was even more unfair than you imagined.   Exponent went beyond the customary bias, spin, and cherry picking of assumptions.  Exponent deceived people about what they had actually simulated and used that to support a conclusion they knew was wrong.

The good news: the problem is simple to show and explain.  The Exponent simulation had balls freely exposed to air to warm up quickly, whereas on game day the balls had remained in a bag, staying cooler.  If you adjust Exponent’s data to reflect the slower warming, their data predicts the pressure that Patriots’ footballs actually had.

All the rest of the Exponent issues are irrelevant because the bottom line is that any reasonable interpretation of their simulation data shows that the Patriots could not have removed measurable air from the footballs.

Folks seem resistant to accepting that Exponent knowingly didn’t simulate what they knew was vitally important.   The proof of it is in the Exponent report if you know where to look.

A “Cognoscenti” op-ed story “DeflateGate, And The Patriots’ False Appearance Of Guilt went on the website 8/31/2015.  The most useful part for proving innocence is this:  “ignoring the fact that the Patriots balls were measured one by one while the remainder of the balls were kept together, cold and wet in an insulated bag that retarded their pressure rise.”  That hyperlink

That link takes you to simple graphics to explain the situation.   It also provides proof in medium and very full detail formats.  It even has the questions to ask Exponent that could have won the appeal by backing Goodell into a position that he knew no court would defend.  This should provide you a practical means to change the minds that matter and get your draft picks back.

Much personal sacrifice went into the work there, as well as into the testing by Mike Greenway that showed that footballs warm dramatically slower in a bag.

Please let me know that you have received this information and provide a least some feedback so that Mike and I can know that we’ve gotten this information to the person most able to take advantage of it.


Robert Young

Simple pictures (to show Exponent knew Pat’s ball pressure was fine)

Link to home page:

Here is the simple explanation as it was when Robert Blecker referenced it in his op-ed piece on WBUR’s web site (a Boston public radio station).

DeflateGate_Key_Issue_In_Pictures r1pt2 updated Sept 13, 2015 (added pic of Patriots bag)
576.6 KiB


Link to home page:

These simple questions (to ask the Exponent witnesses) would have won the appeal: Goodell would have to concede because he’d know he’d lose in Court.   These questions bring out that Exponenet lied about their central conclusion and that Exponent’s research, much quite good, actually proves that no air was removed from the footballs.

The questions weren’t asked because the defense didn’t figure out the “trick” Exponent used because the defense expected bias and spin rather than outright lying.

Key problem: What made it look like the Patriots balls didn’t have enough pressure is that exponent’s simulation of half-time freely exposed the balls to the air to warm up quickly, as compared to on the real game day the balls had been kept in the bag.  The Patriot’s balls had lower pressure than Exponent simulated because the Patriots’s balls were still close to the original field temperature.  Exponent knows this.  Exponent knows that their data proves that no air was removed.

Exponent lied in being unaware that the Patriot’s ball pressure was just right.

These questions would have drawn that out.

1) Based on your research, was the pressure measured in the Patriot’s balls in the locker room at halftime the right pressure* for them to be at if the balls were at field temperature when measured? 

(Correct answer: Yes.) They should readily answer yes.  If they don’t see the supplementary questions.

*”Right pressure”: close enough to be within the band of random variation uncertainty around the measured pressures.  That range is the horizontal shading around the red horizontal line in Exponent Figures 29 and 30.

2.  Did you run a simulation as best as you reasonably could to estimate much pressure the Patriots balls likely gained from warming up from field temperature before they were measured in the locker room at half time?

(Shockingly, the correct answer is no.  What they simulated for halftime was very different from what happened at halftime and they knew it.  They will probably lie by saying yes or playing dumb, so continue to the next question.)

3.  I see that you documented that on game-day the Patriot balls remained in the bag until each ones was measured.  Did you simulate the balls staying in the bag  until each one was measured?

(Shockingly, correct answer is no.  )  Presuming that they play dumb, see the supplemental questions later below).

4. Do think 11 footballs would warm up noticeably slower if they were in a bag, as you documented that they were on game day, instead of spread out in the open?

(If they say no, ask them if they think groceries don’t stay cool better in a bag then out of a bag.)

5. If the ref was right in his recollection of the gauge use, and if you move the expected pressure down a bit from your simulation to reflect even 15% slower warm-up, then doesn’t the actual pressure fit the expected pressure as best as the experiments can discern, meaning that they fit in the uncertainty band around the actual measurement?

(If they don’t agree, use Figure 10 from the Marlow letter or the Amicus brief to prove it).  That figure is reproduced in the supplementary section below.

6. Here is the single most important conclusion your report made:

In both the Non-Logo Gauge and Logo Gauge simulations, …  subject to the discovery of an as yet unidentified and unexamined factor, the measurements recorded for the Patriots footballs on Game Day do not appear to be completely explainable based on natural causes alone
— Exponent report, page 61

You seem clearly aware that the bag is a factor.  You disavowed taking account of the bag when you described the tests.  Your results matched exactly a ball out in the open.  It seems that you must know that keeping the balls in the bag could have slowed the warming.

How can you possibly argue that the statement above was true?


Having proved Exponent lied in their central conclusion, you can also dig deeper (see amicus brief post for proof of much slower warming in a bag, and to put to rest the gauge issue, to show that the Exponent testing completely vindicates the Patriots.  It’s not “uncertain”.  It’s certain that the Patriots did not take air out of the balls.

Follow up questions for if Exponent doesn’t provide the right answers when asked

Follow up for 1:

if they play dumb or say “no”, then direct them to their crucial Figures 29 and 30, follow the dotted “wet ball” curve leftward all the way to the time=0 (which is past the shaded part), then go up a bit to reflect how the red dots are slightly above the dotted line.   It’s in the horizontal shaded band, so its close-enough to the actual game result shown as the horizontal red line).  For more detail, see the Marlow letter or the amicus brief

Follow up for 3:

3a. I see that when you described the first-half simulation, you explicitly mention using bags believed to match the Patriot and Colts bags.  I see you said you replicated the game-day use of the bag.  The bag is mentioned again in carrying the balls to the half-time simulation room.  But then, when it comes to the half-time portion, where it really matters whether the ball is in the bag or not, you say this:

The procedure used to generate the halftime measurements during Game Day was replicated. Namely, the Logo and Non-Logo Gauges were used.
–Source: Bottom of Exponent page 56 (as discussed in my brief on page 35).

Do you agree that the second sentence completely retracts the first sentence, so that all you’re really committing to is using the two gauges, and not replicating anything else? 

Correct answer must be yes.  Google defines “Namely” as meaning “which is to say”.

3b. Did you notice that in your simulations, the balls warmed up at the same speed they warmed up in your earlier lab-bench test with a single ball out in the open on a little pedestal on the table?

(Answer should be yes).

Assuming they say no, point out this:

“The data sets generated by the two methods (game day simulations and the transient curves) correlate well to one another:”
Exponent, Page 59, last paragraph

3c: Did you notice that the transient curves were determined based on a ball alone on a small pedestal on a table.

they noted the correlation between their simulation and the lab-test as if it were a good thing. The data sure does too.  Oops! The data should have showed a lower warming rate, due to the bag slowing down the warming – especially given that it was likely closed the first two minutes before any balls were removed.

5a: see graphic


Prof. Marlow reviewing new-to-him info; may retract support of NFL/Exponent work

Link to home page:

Princeton physics Professor Daniel Marlow reviewed the Exponent report and told the NFL it seemed fine.

8/29/2015: Prof. Marlow sends email 5:15 am confirming receipt.

On 8/28/2015, lunch time, I phoned Professor Marlow to call his attention to the simulation freely exposing balls to the air, vs the balls being in a bag on game-day.  (Recall that is the only reason why the simulation didn’t agree that the Patriots took no air out of the footballs.)

I explained the problem in greater detail.  Although his manner seemed hostile to such a possibility that he might have missed something, he did ask me to send him an email explaining the problem.

I asked that he publicly withdraw his support for the Exponent report.

Below is the email I sent him, only with my home phone removed.

Note: for beyond-a-reasonable-doubt proof that Exponent lied in concluding that their research didn’t explain the Patriot’s ball pressure even if you agree with the ref on gauge use, download may amicus brief. It does that and more.


Better Dialogue #DeflateGate Videos

(Please excuse imperfect production quality of video – -time/skill was limited).

Link to home page:

Proof in the amicus brief proven to far higher precision, with better graphics also, then what is in the video.  Still, the basic concept in the video is exactly the issue.

NFL/Exponent data vindicates Patriots (balls warm slower in a bag) r1 (

Please help #FreeTomBrady (or punish NFL misdeads) by getting this information “out of the bag” for all to see.
Video shows how you can prove from NFL/Exponent’s data that in Exponent’s simulation, what they did was identical to taking the balls out of the bag to warm faster, even though Exponent documented that on game-day the balls were kept in a bag on the floor and removed one by one as they were measured. Re-reading Exponents time-scale to adjust for slower warming vindicates the Patriots. The video also shows Exponent knew their “reasons” disbelieving the ref (about which gauge he used) were senseless.

Supporting docs: see draft Amicus Brief for the Court.
or DeflateGate Home Page

Amicus brief proves beyond doubt Exponent lied in claiming pressure was unexplained.

Link to home page:

10/14: Updated brief posted: better clarity about what really happened at half-time.

9/17: Amicus follow-up letter (downloadable below) now includes proof that the court placed the full amicus brief on the docket.  The court docketed the amicus as part of that letter (document 70) rather than as part of the original request for permission to file.

9/17/2015: Wall Street Journal reporter Jacob Gershman mentioned the brief but not its significance, which is that when adjusting for Exponent having rigged the half-time simulation, Exponents work proves that Exponent cheated and the Patriots didn’t.

Robert F Young Amicus Brief in DeflateGate Aug. 25, 2015 updated 10/14/2015
1.4 MiB
Amicus As Submitted
Amicus As Submitted
amicus as submitted.pdf
1.3 MiB
Amicus Request Letter
Amicus Request Letter
amicus request letter.pdf
101.1 KiB
Amicus Request  Follow Up Letter from court docket.  Proves brief was dockted: NFL says bias/error okay: Response: lying is not.
546.5 KiB

The permission letter may have been sufficient to show the judge a powerful illustration of how lack of due process and lack of fairness truly hurt the Patriots/Brady.  Perhaps this helped sway him in deeming the violations to be material and unacceptable.

The permission letter and follow up letter and now the brief can also be downloaded from court by lawyers admitted to the court, or by using a free PACER account (but if you’re not careful, you’ll run up billable charges for the service).

Below is some additional summary information.

8/28/2015: 7:25pm EDT: Clarified explanation.  Referenced Mike Greenway’s data confirming (prior to my work)  the 75%-slower warm-up that my Amicus brief had predicted based on some simple theoretical calculations.

Exponent’s simulation couldn’t replicate Patriots’ ball pressures because Exponent’s simulation warmed the balls up fast, by freely exposing them to the air.  Proof: Exponent’s “simulation” rate of warming correlating exactly with rate of warming in a previous test Exponent did of one ball at a time, in the open.  Exponent even commented on the correlation (as if it were a good thing, whereas it’s damning to their central conclusion).

In the real half-time locker-room period on game day, the Patriots’ balls stayed in the bag, slowing their warm-up.  Each ball was removed only seconds before it was tested.  Exponent even documented that.

Patriots’ footballs were at just the pressure Exponent’s research said they should be if one accounts for the warm-up being slowed because they were in a bag.

Exponent research vindicated the Patriots.  Exponent lied about how to interpret their research.

This conclusion from Exponent (which is the only one that really matters) is a lie:

In both the Non-Logo Gauge and Logo Gauge simulations, … subject to the discovery of an as yet unidentified and unexamined  factor, the measurements recorded for the Patriots footballs on Game Day do not appear to be completely explainable based on natural causes alone
— Exponent report, page 61.

Exponent used clever wording to sneakily disavow simulating the warm up properly.  In the earlier steps of the “game-day” simulations, they vouched for replicating the use of bags to match game day.  But for the critical part, the halftime locker-room simulation, they used trick wording to commit only to using the same gauges.  There was no lie by Exponent until Exponent feigned inability to explain the Patriot ball pressures.

For a somewhat deeper explanation (but not as deep as in my amicus brief), see my letter to Professor Marlow.  For beyond-a-reasonable-doubt proof, see the amicus brief below.

NFL has argued it can be “wrong” and still be immune from the judge having the power to overturn the decision (source: Rich Levine post: Silly Deflategate has taken a serious turn).

But NFL didn’t argue that they are allowed lie about the fundamental question: was air removed.   Is that their next argument?

For those who say the earlier drafts, or the one erroneously left up on 8/25 at the time Professor Blecker first tweeted about it, the actual one the court put on the docket (added below 8/26, 8:24am) is vastly improved.


Change history:

9/22/2015: Updated the brief.  Again no consequential change.  Technically, it turns out that the conclusion about the Patriots-Colts pressure differential, which was always a moot point relative to the question of whether the Patriots’ balls had the right pressure, was an intentional deception but not a lie.  Legally speaking, the final conclusion of Exponent (item 13 on page 68) never claimed that the patriots-colts pressure differential was inexplicable.  Instead, legally, it only claimed that the source of the explanation did not lie within the in-game events.  Of course it didn’t: the explanation was the combination of the difference in pressures when measured by the ref pre-game, and especially in the different handling of the Patriots and Colts balls during the half-time period in the locker room.  The locker room period (rather than the game events) provides the explanation that had seemed missing.

9/11: Noted that the Judge posted my letters, and that the letters may have had an impact.

9/9: Updated title of post to reflect current use of the info.  Moved download links to the top and posted minor correction to the brief.

8/27: Tests of actual balls warming up (slowly) in a bag by Mike Greenway, found on his website, in his “Exponents (sic) fundamentally flawed research” link as of 8/28/2015 6:56PM  EDT (dated July 19) confirms nicely the 4x slower in-bag warming I predicted theoretically in the amicus brief.  As of the time noted, I disagree with some things Mike says is wrong with the Exponent report.  I contend that the Exponent data and research is good, and that it just needs a reasonable adjustment to the simulation to reflect the lower warming rate on game-day as compared to Exponent’s simulation. Result: Exponent’s own data vindicates the patriots.   Credit due: Mike did identify the in-bag issue (before I noticed it on my own) and he even did a test!!!! I believe I was unique in identifying the accurate way to use this to show Exponent lied in their conclusion, and to further use the Exponent data to prove that the Patriots did not remove air from the footballs.   I believe I’m also unique in totally shredding the “ref was wrong on the gauge” argument by the NFL (proving that Exponent believed none of the reasons Exponent gave against the ref) and to show that had Exponent changed Exponent’s Colts-timing assumption to better match reality, Exponent’s own data affirmed that the ref was right about the gauge use.

8/26/2015: 7:50pm: court says judge not considering any (contrary to Judge’s clerk earlier in day)

8/26/2015: 11:10am: Overnight mailed follow up letter.  Responds to new NFL contention that bias or getting facts wrong is not grounds to overturn.  Response: even if so, colluding with Exponent to lie on the central issue is grounds to overturn.  Also called out the trick wording that was omitted from the original request letter for the sake of brevity but was discussed in the brief page 35 and 36.

8/26/2015 9:10am: Court confirms judge will review request letter submitted with the actual brief.  Hopefully he will also read the brief.



Better Dialogue re: #DeflateGate (NPR radio station) 8/31/2015 Op-ed “DeflateGate, And The Patriots’ False Appearance Of Guilt references this site.  The the author of that piece (Robert Blecker) recommends particularly the third link below:

Analysis of the Exponent report indicates that the headline from the investigation would have been: “Patriots right on their science: no evidence of wrongdoing” , except that, to please the NFL, Exponent applied assumptions they knew to be wrong and misrepresented how their simulation compares to the real-world events.  Absent those assumptions, the rest of the Exponent experiments and analysis thereof strongly supports the Patriots.  Thus #FreeTomBrady

The analysis is in the Amicus brief  that was submitted to the court for consideration, received 8/26 before 10am.    It relates to the court case and not just the data.

Attention-catching highlights from the above brief (I will add convenient links to details later if there’s enough traffic)

  • Basically, the only reason the Exponent data is incriminating is that Exponent’s simulation didn’t keep the balls in the bag during their simulated halftime like the NFL did in January.
  • When reading Exponents charts, only this time adjusting for the balls’ warming slower in a bag, Exponents data clears the patriots of wrongdoing.
  • None of the reasons Exponent gave for not believing the ref (about what gauge was used) were ones Exponent actually believed made sense; they don’t.
  • The NFL is hurting the integrity of the game with new rules that pump up balls by 2 extra psi for the 2nd half of 30-degree games for 2015 as compared to all previous years.

Best next steps to verify for yourself: See the list of links at the top of this post.  use the above Amicus brief link for deeper proof, beyond a reasonable doubt.


Sorry for crude website formatting.

Please submit only comments addressing specific steps in the arguments proposed below.  I’ll correct my analysis and conclusions if needed.


9/26: Exponent never claimed pressure difference unexplained; lawyered up language fooled lawyers.  Additional letter to NY times.

9/18, 9/21: New York Times op=ed request/correction request and follow on request.

9/17: Wall Street Journal mentions the amicus brief but not its importance.

9/16: A reporter for a major national newspaper called me and informed me that the amicus brief I submitted is now on the official court docket.

9/11: Court put my letters on the docket

9/9: added the Amicus brief link to the bullet list.

8/31: Reference the editorial (early AM), Open letter to Robert Kraft (9:13 am)

8/30: simple picture explanation added

8/26: see hot/new.

Aug. 20 (just after midnight)

  • Brief improved (see brief point),
  • Above points simplified: greater clarity of the real point although loss of precise numbers.  For numbers see the brief