A Skeptical Audience

So Donald Trump fired FBI director James Comey, as everybody knows by now. This immediately drew parallels to the Saturday Night Massacre, where Nixon ordered his Attorney General to fire special prosecutor Archibald Cox.

At the heart of the matter are three questions. First, whether Comey was the best person for the job. The second question is whether this will hinder any ongoing investigations, specifically the investigation into Russian involvement in the Presidential election; and what if anything will be done to safeguard these investigations.

The third question is the most important, and is simply; which of the first two questions did Trump answer when firing Comey? That is, did he fire Comey because he felt Comey wasn’t a good director? Or did he fire Comey specifically to stop the Russian investigation (or any other)?

If it’s the first, and Comey wasn’t the best fit for the job (or even if Trump just honestly thinks so), then there’s no real problem. If it’s the second, then basically Trump is abusing the office to enact a personal agenda, using the department of justice as a political tool, and obstructing justice. Essentially, what Trump did is either a standard (if somewhat unusual) way of acting as President, or an impeachable offense; either it was as bland as Bill Clinton dismissing FBI director William Sessions, or as corrupt as Nixon firing Cox.

* * *

Having just finished a class on negotiation; I was struck by an study by Huthwaite inc, called “The Behaviors of Successful Negotiators” which looked at (no surprise here) the behaviors of successful negotiators. It was one of those truly great readings, the point of which seems obvious to me after reading it, but never once occurred to me beforehand. There were some behaviors which were kind of boring, (such as skilled negotiators talking more about long term topics), some which were somewhat reasonable but not much use elsewhere (skilled negotiators don’t have a predetermined sequence of when they wish to discuss which issue), but some which were meaningful and profound.

The fact I was most impressed with were the fact that skilled negotiators rarely used words like “generous,” “fair,” or “reasonable” to describe their own offers. Within a negotiation, calling something “fair” which you present, even if you think it is fair, is unlikely to convince your counterparty; it will only serve to aggrevate him or her if they don’t think its fair.

The second fact which I thought was really valuable was that skilled negotiators often made fewer arguments in their favor; but those were typically better quality. That is, if you’re trying to convince a hostile (or even skeptical) audience of a fact, and you present 5 arguments, they’re naturally going to latch on to the weakest argument you make. People are not rational Bayesian calculation machines, 2 strong arguments in favor are greater at convincing humans than 3 strong and 3 weak arguments in favor.

In both cases, the lesson is similar, in convincing a skeptical audience less is often more. Using more neutral language is more persuasive than persuasive language, and using fewer arguments is more persuasive than using many. The stronger you believe something (or come across as believing that thing), the less persuasive you can be to a skeptical audience.

* * *

Today, while packing up my apartment, I was listening to a several months old episode  of Bill Simmons’ podcast, the BS report. Simmons’ guest was one of my favorite essayists, Chuck Klosterman. In it, they were discussing who the NBA MVP will be, Russell Westbrook or James Harden. Klosterman had a very intriguing argument, which essentially went like this (I am of course paraphrasing):

1: The people who favor Westbrook think its obvious that the MVP should be Westbrook and there’s no other choice. (Westbrook averaged a triple double over the season, something that hasn’t been done since the legendary Oscar Robinson).

2: The people who favor Harden think its an interesting question and there are arguments for multiple candidates, but Harden is overall their favorite.

3: The people who are undecided will see the above two arguments, and almost tautologically will relate to 2nd one more. Because (by definition) they haven’t made up their mind, they can relate to those who think its a close race, who favor Harden.

Its an almost brilliant idea, and I have no idea if it will be correct or not, but it has a certain logic to it. Underlying it all is the same lesson, your ability to convince a skeptical party of something can be inversely related to the strength of your own belief.

* * *

Trump’s approval rating is low, but its not historically low. 538.com has him at 41%; Gallup had Obama at the low 40’s for much of 2011 and 2014 (of course this time in 2009 Obama was in the low 60’s, 20 points better than Trump is now.) Gallup had W Bush’s low at 25%, HW Bush at 30%, Clinton at 40%, Reagan at the mid 30’s, Carter at 30%, and Ford at 40%. (the counterclaim is that none of these Presidents were this unpopular this soon in their Presidency, but that’s kind of beside the point).

I’m not a Trump fan, and I certainly don’t think I will be. I see a number of things which Trump has done as being bad, corrupt, or incompetent, and the demeanor in which he has conducted himself has at times seemed unhinged and almost crazy. This to go with his numerous scandals, problems, gaffes, and remarks he made while campaigning and as a public figure.

Yet as much as 41% of American voters still approve of Trump. Why is this?

* * *

All this brings us back to President Trump and James Comey. If you’re already inclined to believe that Trump is a despot you will probably see the Comey situation as analogous to the Saturday Night Massacre, and Trump as obstructing justice. If you’re a fan of Trump you’re much more likely to see the situation as nuanced, or as analogous to Bill Clinton dismissing Director Sessions. And if you’re in between? Well, articles like this probably won’t convince you.

I think that those people who either support Trump or at least still giving him the benefit of the doubt just see all the criticism blending together, drowning itself out. They see Trump detractors reacting to firing Comey in the same way the reacted to Trump’s February 16 press conference. As long as they see the tone and not the substance of their opponents’ arguments, they’ll get no new information, and of course won’t change their minds.

Take an article like this, very anti-Trump, which purports to list all the bad things Trump has done. Yet it seems like half the things listed aren’t things Trump has done, but rather things he’s said or tweeted. Going back to the fact about skilled negotiators, how they will use fewer stronger arguments. Then compare that to the list in the nymag article. For many of the “Trump said this” arguments, if you’re not convinced now, you may not be ever. And if that argument won’t convince a skeptic, then making it will probably make better arguments less convincing.

With a skeptical audience, the strength of your beliefs can often work against you. There’s no shortage of liberal antipathy towards the Trump administration, yet I wonder if the strength of the left’s beliefs is actually hindering its ability to make a convincing argument.


All time greats

Quick Question, who are the best basketball players since 1980?

In semi-reverse chronological order they are:

Lebron James
Kobe Bryant
Tim Duncan
Shaquille O’Neal
Michael Jordan
Hakeem Olajuwon
Isiah Thomas
Magic Johnson
Larry Bird
Kareem Abdul-Jabar

You can argue that Isiah shouldn’t be on there; or that, Barkley, Pippen, and either Karl or Moses Malone should be, and maybe Dirk or KG, (also you can argue for Durant or Anthony David should be added, but to that I’d say wait). I wouldn’t agree with displacing any of those players (besides Isiah) though. If you want to get technical, you might argue that if you only take his play since 1980, Kareem doesn’t quite belong on the list, but a: you’re wrong, he does, b: who cares and c: its pretty much irrelevant since he was the best damn player on the planet throughout the 70’s.

Overall, its a damn good list; specifically, it’s the list of players (in my judgment) who were the best player on at least two championship teams. It does a remarkable job of filtering the great from the good. At least for a (somewhat) objective standard; it matches up really really well with subjective opinion, certainly more than any other stat I can come up with.

If we sort the list based on number of championship teams they were the best player on, we get:

Michael Jordan: 6 (’91, ’92, ’93, ’96, ’97, ’98)
Tim Duncan: 4 (’98, ’03, ’05, ’07)
Shaquille O’Neal: 3 (’00, ’01, ’02)
Magic Johnson: 3 (’82 ’87, ’88)
Larry Bird: 3 (’81, ’84, ’86)
Lebron James: 2 (’12, ’13)
Kobe Bryant: 2 (’09, ’10)
Hakeem Olajuwon: 2 (’94, ’95)
Isiah Thomas: 2 (’89, ’90)
Kareem Abdul-Jabar: 2 (’80, 85)

Again, almost a perfect list. Tim Duncan is too high (in terms of history to the game, he should be closer to Kobe and Hakeem than Jordan), and obviously Kareem is too low (but he rises to three if you count his title in 1971, which puts him next to Magic and Bird, which is pretty much right where he belongs). I guess you could argue that Lebron is too low, but he still has time. Overall though, this is a really good proxy of who was the best in the NBA since 1980; not perfect, but I doubt you could find a semi-objective stat which does better.

If we apply this throughout the history of professional basketball, we add:

Wilt Chamberlain
John Havlicek
Bill Russell
George Mikan

Ok, not as good of a list, as we’ve got legitimately great players we’re leaving off, most notably Oscar Robertson and Jerry West. You can argue that Havlicek doesn’t actually qualtify (ie, that Cowens and White were better in ’76, and Russell was better in ’69 and every preciding year they won the champinoship). I disagree, Havlicek was the best Celtic in ’74, and I’d argue in ’69, a case can be made for ’68 as well).

You can also argue that Wilt wasn’t the best player on the ’72 Lakers (that title belongs to Jerry West), therefore taking him off, but I think this just goes to show that the list doesn’t work as well back during the period when Bill Russell and the Celtics won every year.

Obviously George Mikan doesn’t belong in the same discussion as Magic and Jordan, he gets here on the virtue of being the best player in the NBA before the NBA was important. Think of him as you would Cap Anson or Dan Brouthers in baseball, which is to say not at all.

If you include ABA titles, then Dr. J joins the club. Finally, you can make the case that either Willis Reed or Walt Frazier belong here (but not both), although I have Reed as being the best on the ’70 Knicks with Frazier being better in ’73, but I think you can make the case for either player in either year. Besides, the fact that there is a case to be made means taht neither player was clearly the best player on two championship team.

Overall, this is a really interesting phenomena; if you make adjustments for Oscar and West (who probably would have won more titles had the Celtics not dominated the 60s), then you get a really damn good list of the best players of all time.

Putting everybody in order:

Russell 10
Jordan 6
Mikan 5
Duncan 4
Shaq 3
Kareem 3
Magic 3
Bird 3
Kobe 2
Lebron 2
Wilt 2
Hakeem 2
Havlicek 2
Isiah Thomas 2
Julius Earving 2 (ABA titles)

Reduce Russell a few spots, remove Mikan altogether (for justifiable reasons), drop Duncan a few spots, drop Shaq one and bump up Wilt, add Oscar, add West, and remove Isiah. That might not only be the list of the best basketball players ever, but in order! (Ok, so that seems like I’m cheating; but there are two things to consider; first I will make no apologies for removing Mikan: he played before the shot clock in a mostly white league. Second, half those adjustments are adjustments to the fact that 1960’s Celtics won so many titles, improving West, Oscar and Wilt while reducing Russell. So in the end there’s really only four changes you make in order to get an almost perfect list; adjust Duncan down, remove Isiah, drop Shaq one, and adjust for the 60’s Celtics.  And its not like any of these are horrible errors, I don’t think Isiah is one of the 15 best players ever, but he’s in the top 25)

You may also say I’m cheating based on how I define the best player, after all there is some subjectivity here. But there’s little actual effect here; most seasons either: 1: had a clear dominant player on the championship team, 2: had unclear dominant players on the championship team, but which didn’t effect this analysis (ie, Detroit in ’04, Boston in ’08), or had a pair of players who contend for best player in the championship team but who are both on this list (Kareem/Magic, Shaq/Kobe).

This seems incredible to me, that there is a mostly subjective way to measure something in sports that seems to work almost perfectly. Also, the way it works seems to be flexible as well, let me explain. Lets return to Havlicek; he’s on the border of all-time great. No one would doubt he was a hall of famer and incredible player, its whether his comparables are more to someone like Magic or more like Rick Barry. So we’ve got a metric, which could go either way on Havlicek, and opinion is that… he is on the line between in and out of all time.

Or look at Julius Erving. How you view him historically is really a question of how you view the ABA. If you think it was comparable to the NBA, then you list him as an all-time great, if you don’t, then he’s merely a hall of famer. This lines up exactly with our metric, if you include the ABA titles, then he’s in the discussion with Bird and Magic, if not, then you don’t.

Compare this to football, where it seems that any subjective and simple rankings of QBs would either give you Bradshaw over Peyton (which doesn’t seem right to me), or something like Favre or Marino over Montana (which seems awful). Yeah, you could weight things (a passing title worth X, a superbowl worth Y) but then you’re no longer simple (in this case there may be no subjectivity once you create the model, but the number of ways you can tinker with the model are endless, so you can get any remotely reasonable ranking).

Now this would be a really great essay if I could somehow relate this some larger point about society or something, but alas I cannot. But more than anything else, I think it relates to the way that basketball is played, how unlike in baseball or even football, one player can dominate a game. And finally, it gives us a clear method of determining goals for future basketball greats. How can we tell if Durant will join the club? Simply, if he leads his teams to two championships.