How will Trouble do in September?

[marcus]

How will "Trouble" do in September?

"The Trouble with Physics...and What Comes Next" has been out for almost two years now. It started shipping from Amazon in August 2006 (before the official September date).

It is the only popular book I know that discusses the progress being made in non-string Quantum Gravity (QG) research---and describes some of the various QG approaches. So how the book sells, compared with popular stringy books as a benchmark, is an indicator that bears watching.

I wish there were other indices to watch besides this one. Certainly the appearance in the July Scientific American of a fine article by members of the Utrecht QG group is a hopeful sign. Here's the link:
http://www.scribd.com/doc/3366486/SelfOrganizing-Quantum-Universe-SCIAM-June-08
And there are other indicators like conference participation, research publication rates, citation numbers, projected faculty hiring and so forth. But no other index of public awareness.

The bottom line is that there seem to be changes occurring both in comparative public recognition and research emphasis as regards string and non-string approaches to the major problems in fundamental physical theory. I guess one can call them sociological changes. And one way of keeping track is to see how this book does.

So we calculate the ratio of "Trouble" salesrank to the average salesrank of the five currently most popular stringy books. If the ratio is 1.0 that means that Trouble sales are roughly on par with benchmark. If the ratio is 2.0, then judging by salesranks Trouble is doing twice as well as benchmark. If the ratio is 0.5 it is doing half as well. And I would say that if it goes down to 0.1 that would mean the book has (at least temporarily) dropped out of sight.

So how do you think Trouble will be doing this September 1?
Here's the record since September 1, 2007. Readings taken at noon pacific time.


1 September 6.4
1 October 6.5
1 November 5.2
1 December 2.4
1 January 1.5
1 February 1.3
1 March 0.4
1 April 0.6
1 May 1.0
1 June 1.0
1 July 0.5
5 July 1.0
6 July 1.0
...
...
1 September ??


Here's a link to the previous forecast poll's conclusion. We had four competing predictions. Mormonator obviously should have won but there was a last minute upset

right on the button, correct to the nearest two decimal places---the ratio at noon was 0.504
...

 

[marcus]

To give an example of how the ratio is calculated, I'll show it for 7 July.


...
1 June 1.0
1 July 0.5
5 July 1.0
6 July 1.0
7 July 1.3
...
...
1 September ??


"Trouble" salesrank was 3140.
the five most popular stringy books that day, and their salesranks were
elegant 2744
parallel worlds 2786
fabric 3837
susskind's book 4626
elegant (hardbound) 5885
for an average of 3975.6

so the ratio was 3975.6/3140 = 1.3

Roughly this means that, judging by salesranks, Trouble was doing slightly better than par compared with the string benchmark. Its salesrank was roughly equal, or slightly better, than the average we use for comparison.

In fact, par has been typical for a good part of of June and first week of July---although unusual circumstances did cause a brief drop in the index right around the first of July. This on-par standing, recently, has persuaded me to guess that the Trouble salesrank will not dwindle away by the end of summer, as I earlier expected, but may hold to around par for a few more months.

The book has already had a rather long market life. It has been out for almost two years, during which several other wide-audience books about fundamental physics have entered the market and then largely disappeared from view.

 

[marcus]

1 September 6.4
1 October 6.5
1 November 5.2
1 December 2.4
1 January 1.5
1 February 1.3
1 March 0.4
1 April 0.6
1 May 1.0
1 June 1.0
1 July 0.5
...
11 July 0.6
...
...
1 September ??

After several days before and after 1 July of bouncing around 1.0 the index has again subsided to near 0.5. I feel less optimistic now than I did earlier about its prospects for September. Mormonator's forecast could well prove right.

 

[marcus]

not dropping out of sight just yet

1 September 6.4
1 October 6.5
1 November 5.2
1 December 2.4
1 January 1.5
1 February 1.3
1 March 0.4
1 April 0.6
1 May 1.0
1 June 1.0
1 July 0.5
...
12 July 1.4
13 July 1.3
...
1 September ??

Smolin's book has had an odd sales history. It doesn't seem to want to drop out of sight (as several other physics books for general audience have after a year-or-so's prominence). Something always seems to stir up further interest and revive sales. It may be percolation through expanding demographic layers as Mormonator suggested earlier. It could also be that new books coming on the market ignite interest in this one.

At noon on 12 July "Trouble" salesrank was 3846 and the stringy top five average was 5225.0. On 13 July "Trouble" stood at 2488 and the top five benchmark was 3179.8. A new book about string, "The Complete Idiot's Guide..." is consistently among the five most popular string titles these days.

I confess to being quite puzzled by the way the index has been jumping around lately, but I haven't given up on my guess that it will still be around 1.0 in September.

 

[marcus]

1 September 6.4
1 October 6.5
1 November 5.2
1 December 2.4
1 January 1.5
1 February 1.3
1 March 0.4
1 April 0.6
1 May 1.0
1 June 1.0
1 July 0.5
...
12 July 1.4
13 July 1.3
14 July 2.1

...
1 September ??

book seems to have 9 lives, like the cat who keeps coming back.
remarkable resiliency, at noon 14 July (Bastille Day) its rank was 2488 and the five most popular stringies were elegant,fabric,parallel,complete idiot's guide, and hyperspace---their average rank was 4321.6. So the ratio was 4321.6/2488 = 2.1
I almost didn't put 2.0 on the poll, as a possible choice, thinking that from here on it would never get that high

 

[marcus]

1 September 6.4
1 October 6.5
1 November 5.2
1 December 2.4
1 January 1.5
1 February 1.3
1 March 0.4
1 April 0.6
1 May 1.0
1 June 1.0
1 July 0.5
...
19 July 1.1
20 July 1.0

...
1 September ??

Uneventful days. Yesterday noon the benchmark was 3989.6 and Trouble's rank was 3520.
Today benchmark was 3862.8 and Trouble was at 4033, about par with the stringy average.

Or could current excitement about Garrett Lisi (the 21 July 2008 New Yorker piece) be artificially stimulating sales which would otherwise have been sagging? Smolin's book helps tell the story of several different approaches to quantum modeling spacetime which are interesting and may deserve support, besides dominant string. So, if Lisi's gambit can be taken as an example of an alternative approach unexpectedly coming to light, then it has a kind of resonance with Smolin's book.

 

[mormonator_rm]

1 September 6.4
1 October 6.5
1 November 5.2
1 December 2.4
1 January 1.5
1 February 1.3
1 March 0.4
1 April 0.6
1 May 1.0
1 June 1.0
1 July 0.5
...
19 July 1.1
20 July 1.0

...
1 September ??

Uneventful days. Yesterday noon the benchmark was 3989.6 and Trouble's rank was 3520.
Today benchmark was 3862.8 and Trouble was at 4033, about par with the stringy average.

Or could current excitement about Garrett Lisi (the 21 July 2008 New Yorker piece) be artificially stimulating sales which would otherwise have been sagging? Smolin's book helps tell the story of several different approaches to quantum modeling spacetime which are interesting and may deserve support, besides dominant string. So, if Lisi's gambit can be taken as an example of an alternative approach unexpectedly coming to light, then it has a kind of resonance with Smolin's book.

This is bizzarre behavior indeed, though the explanation isn't too unexpected. The resonance idea could prove true in some respects. The idea that buyers of the latest alternative-approach type of work will rehash the previous iterations and, hence, boost sales of the previous works of interest isn't too hard to believe. So, your theory that Lisi's work coming to light is causing some "resonance" behavior with Smolin's is not necessarily out of the ball-park... After all, this effect has been seen in consumer markets of other types, such as with movies of like plot, books of like plot, and even my favorite, unlocked cell phones of similar variant. So there you go!

 

[marcus]

...The idea that buyers of the latest alternative-approach type of work will rehash the previous iterations and, hence, boost sales of the previous works of interest isn't too hard to believe. So, your theory that Lisi's work coming to light is causing some "resonance" behavior with Smolin's is not necessarily out of the ball-park...

It was just a thought. Probably neither of us found it very convincing. Here is the latest:

1 September 6.4
1 October 6.5
1 November 5.2
1 December 2.4
1 January 1.5
1 February 1.3
1 March 0.4
1 April 0.6
1 May 1.0
1 June 1.0
1 July 0.5
...
19 July 1.1
20 July 1.0
21 July 0.7
22 July 0.9

...
1 September ??

It could be that the thing is being artificially supported by the Lisi New Yorker business in which case we will see in a week or so---after the stir dies down our book's index will drop off. But it could also just hanging in around par of its own accord.

 

[marcus]

In case anyone is interested what the numbers are like these days, I just took an informal tally at 7PM (my usual time is noon) and Trouble salesrank was 4029. The top five stringy books were
black hole war 579
fabric 2587
elegant 2889
idiot guide 4304
elegant hardbound 5612
so the average was 3194.2 and the ratio was 3194.2/4029 = 0.8

 

[marcus]

registering a new low

1 September 6.4
1 October 6.5
1 November 5.2
1 December 2.4
1 January 1.5
1 February 1.3
1 March 0.4
1 April 0.6
1 May 1.0
1 June 1.0
1 July 0.5
...
23 July 0.4
...
1 September ??

today at noon
black hole war 1034
fabric 2425
elegant 2694
idiot's guide 4078
elegant hardbound 6377
average 3321.6
Trouble's rank 7568, and ratio 3321.6/7568 = 0.4

Earlier I was feeling pretty confident of winning on the September 1 forecast, but now it looks as if Mormonator has a palpable chance.

 

[marcus]

Trouble with Physics is getting blasted by the stringy competition. Again today at noon the ratio was 0.4. In case anyone wants to see the numbers for 24 July:

trouble 9183

compared with the benchmark average 4016.6 of the following top five
black hole 1531
elegant 3455
parallel 3713
fabric 4790
idiot guide 6594

 

[marcus]

another new low---this may be the final plunge

1 September 6.4
1 October 6.5
1 November 5.2
1 December 2.4
1 January 1.5
1 February 1.3
1 March 0.4
1 April 0.6
1 May 1.0
1 June 1.0
1 July 0.5
...
23 July 0.4
24 July 0.4
...
26 July 0.3
...
1 September ??

today 26 July at noon pacific the ratio was 0.3
To my knowledge, this is the lowest it has been in the approximately 2 years that Trouble has been on the market.
It has done well. During the first year it was often times at the top of the physics bestseller list, or in the top three. Much of the time the ratio we've been watching has been 2 or 3 and better (twice or thrice the stringy topfive benchmark). But popular mass market books often do have an arc like this, where they get replaced by something else and drop out of sight. I think that is happening now.

Well it came out in August 2006. Officially in September, but actually was in stock and shipping already in August. So it's had a good two year run and done way better than most people expected, I think.
Mormonator's estimate may in fact be considerably too high, as it turns out---although still the closest on record.

 

[Astronuc]

I just bought a copy of Trouble along with Roger Penrose's Road to Reality, so maybe there will be a jump in the sales this July. I also bought Gell-Mann's The Quark and the Jaguar.

 

[marcus]

I just bought a copy of Trouble along with Roger Penrose's Road to Reality, so maybe there will be a jump in the sales this July. I also bought Gell-Mann's The Quark and the Jaguar.

Delighted to hear it! The index was encouraged by the news and perked up some today.
It was 0.6 (still under par but not so badly as a couple of days ago).

Roger Penrose's book is packed with mentally challenging stuff that is so clearly explained that you feel you ought to understand it. Anybody who reads thru even half of that book deserves respect and admiration by my standards. Glad you intend to try it.

I don't know The Quark and the Jaguar, it will be interesting to hear your impressions of all three books.

 

[marcus]

1 September 6.4
1 October 6.5
1 November 5.2
1 December 2.4
1 January 1.5
1 February 1.3
1 March 0.4
1 April 0.6
1 May 1.0
1 June 1.0
1 July 0.5
...
30 July 0.5
...
1 September ??

Smolin book is taking a drubbing. Only doing half as well as the stringy topfive average that serves as our benchmark. Today 30 July at noon Trouble salesrank was 5455 compared with
Black hole 111
Parallel worlds 2300
Elegant 3041
Elegant hardbound 4043
Fabric 5289
making the benchmark average 2956.8 and the ratio 0.5

One month to go before our forecasts' target date.

 

[marcus]

1 September 6.4
1 October 6.5
1 November 5.2
1 December 2.4
1 January 1.5
1 February 1.3
1 March 0.4
1 April 0.6
1 May 1.0
1 June 1.0
1 July 0.5
1 August 0.4
...
1 September ??

One month to go before our forecasts' target date.

 

[marcus]

Still showing signs of life.

As of noon pacific on 2 August
the Smolin book's salesrank was 2739
and those of the five currently most popular stringy books were
black hole 208
elegant 2896
fabric 2899
parallel 4946
idiot's guide 5131
making the benchmark average 3216.0 and the ratio 1.2
So "Trouble" was standing slightly better than par compared to benchmark.

 

[marcus]

As of noon pacific on 3 August
the Smolin book's salesrank was 2890
and those of the five currently most popular stringy books were
black hole 226
elegant 1993
fabric 2541
parallel 2905
idiot's guide 4083
making the benchmark average 2349.6 and the ratio 0.8
"Trouble" was standing slightly below par compared to benchmark.

 

[marcus]

Another brave soul has joined us! So we now have three different predictions for the salesrank of "Trouble" relative to benchmark, as of 1 September 2008. Welcome to Ed Aboud! Here is the record of the past year:


1 September 6.4
1 October 6.5
1 November 5.2
1 December 2.4
1 January 1.5
1 February 1.3
1 March 0.4
1 April 0.6
1 May 1.0
1 June 1.0
1 July 0.5
1 August 0.4
2 August 1.2
3 August 0.8
4 August 0.5
5 August 1.2
...
1 September ??


On 5 August, shortly after noon pacific time, the Smolin book's salesrank was 2963
and those of the five currently most popular stringy books were
black hole 518
fabric 3951
elegant 4172
parallel 4291
idiot's guide 4945
making the benchmark average 3575.4 and the ratio 1.2
"Trouble" again slightly above par compared to benchmark.

 

[mormonator_rm]

Another brave soul has joined us! So we now have three different predictions for the salesrank of "Trouble" relative to benchmark, as of 1 September 2008. Welcome to Ed Aboud! Here is the record of the past year:


1 September 6.4
1 October 6.5
1 November 5.2
1 December 2.4
1 January 1.5
1 February 1.3
1 March 0.4
1 April 0.6
1 May 1.0
1 June 1.0
1 July 0.5
1 August 0.4
2 August 1.2
3 August 0.8
4 August 0.5
5 August 1.2
...
1 September ??


On 5 August, shortly after noon pacific time, the Smolin book's salesrank was 2963
and those of the five currently most popular stringy books were
black hole 518
fabric 3951
elegant 4172
parallel 4291
idiot's guide 4945
making the benchmark average 3575.4 and the ratio 1.2
"Trouble" again slightly above par compared to benchmark.

I made sure to order the library a copy of "Trouble" off of Amazon before my departure. I hope I helped the salesrank...

 

[marcus]

I made sure to order the library a copy of "Trouble" off of Amazon before my departure. I hope I helped the salesrank...

A noble but self-defeating gesture, Mormonator. Yours is the low estimate in the poll, you should be doing everything possible to reduce sales!

 

[mormonator_rm]

A noble but self-defeating gesture, Mormonator. Yours is the low estimate in the poll, you should be doing everything possible to reduce sales!

Well, what can I say... the nobility of self-sacrifice...

LOL

 

[marcus]

actually you are doing fine, as we both know
The index was 0.6 today, when I checked at a few minutes past noon.
So if today were target it would be a tie between you and Ed Aboud.

You know what would make a great popular Quantum Gravity book would be the
triangulations path integral approach that Jan Ambjorn and Renate Loll are developing
where they simulate the emergence of small universes in the computer.
Did you see the July 2008 issue of SciAm, their article?
In case not, I have a link to it in my signature----the scribd.com one.
Officially the approach is called CDT---they explain that bit of terminology.
More CDT graphics are in this set of slides for a presentation at a recent conference.http://echo.maths.nottingham.ac.uk/qg/wiki/images/1/1e/GoerlichAndrzej1214824381.pdf
I was impressed by Goerlich's slides and started a thread about them here
https://www.physicsforums.com/showthread.php?t=249037

 

[marcus]

1 September 6.4
1 October 6.5
1 November 5.2
1 December 2.4
1 January 1.5
1 February 1.3
1 March 0.4
1 April 0.6
1 May 1.0
1 June 1.0
1 July 0.5
1 August 0.4
...
9 August 0.9
...
1 September ??


As of noon 9 August, TWP salesrank was 5291
and ranks of the five currently most popular stringy books were
black hole 682
elegant 3866
fabric 4855
parallel 5514
idiot's guide 9468
making the benchmark average 4877.0 and the ratio 0.9
slightly below par.

 

[marcus]

16 August 0.7

the noon ratio has been fairly stable and the five currently most popular stringy books are the same as they were a week ago, in the same order: hole, elegant, fabric, parallel and idiot's guide.

If today were the first of September, Ed Aboud would be the winner, and if I were to guess at this point what the ratio will be on 1 Sept my forecast would echo Ed's.

I don't know if we can see any obvious trends at the moment. One thing to notice is that the black hole book is not primarily about string theory---the stringy stuff comes in the last 1/4 towards the end. The book is primarily a memoir about the black hole information controversy (which is not yet actually settled, as the reviewer in Nature pointed out.) It has consistently been among the top five most popular stringers (if viewed as such) ever since 1 August when it came out.

 

[Gokul43201]

Since day-to-day variability seems to span 4 choices in the poll, I recommend some kind of average (say, an unweighted mean of daily values from Aug 26 through Sep 5) be used for a determination of the value of the Sep 1 ratio.

 

[marcus]

Since day-to-day variability seems to span 4 choices in the poll, I recommend some kind of average (say, an unweighted mean of daily values from Aug 26 through Sep 5) be used for a determination of the value of the Sep 1 ratio.

Good! I was thinking essentially the same thing but didn't know how many days on either side of target to take----how long a duration to give the window.

Just to make it a little less work, and achieve roughly the same effect, I think I will take a 7 day window (instead of the 11 day window you propose). the unweighted average of

29 August
30 August
31 August

1 September

2 September
3 September
4 September

(BTW enviable move--- a great place. not to mention several top inst higher learning, there is all that cultural stuff. even living in SF bay area I experience some wistfulness when I talk to people from there. Congrats.)

================

Will switch to the new system as of this time, unless Ed Aboud or Mormonator object.

Today the noon ratio was 0.5 and the five most popular stringy books were black hole, fabric, elegant, parallel and idiot's guide, in that order.

 

[marcus]

Unexpected spike in the noon ratio. Maybe there was some media event: debate, interview,...


1 September 6.4
1 October 6.5
1 November 5.2
1 December 2.4
1 January 1.5
1 February 1.3
1 March 0.4
1 April 0.6
1 May 1.0
1 June 1.0
1 July 0.5
1 August 0.4
...
16 August 0.7
...
18 August 1.4
...
1 September ??


As of noon 18 August, TWP salesrank was 3927
and ranks of the five currently most popular stringy books were
black hole 323
parallel 4038
elegant 5153
idiot's guide 8878
fabric 9071
making the benchmark average 5492.6 and the ratio 1.4

Computing the 1 September ratio as a simple average of seven readings around that date, along the lines suggested, definitely appeals. Unless I hear objection in the next couple of days, I'm going to do it that way from now on. That will iron out such spikes and wild excursions as we've been seeing recently.

 

[marcus]

OK I haven't gotten any objections, so it's official. I will take Gokul's suggestion and define the 1 September ratio as the average of the noon ratios for 7 days right around that date. this will get rid of some of the fluctuation which we have been seeing.

TWP has been staying consistently around par recently


1 September 6.4
1 October 6.5
1 November 5.2
1 December 2.4
1 January 1.5
1 February 1.3
1 March 0.4
1 April 0.6
1 May 1.0
1 June 1.0
1 July 0.5
1 August 0.4
...
...
18 August 1.4
19 August 1.0
20 August 0.9
...
1 September ??


As of noon 19 August, TWP salesrank was 6004
and ranks of the five currently most popular stringy books were
black hole 1118
parallel 4485
elegant 5569
elegant hardbound 8086
idiot's guide 10,623
making the benchmark average 5970.2 and the ratio 1.0

As of noon 20 August, TWP salesrank was 6192
and ranks of the five currently most popular stringy books were
black hole 1450
fabric 4039
elegant 5555
elegant hardbound 6281
parallel 9331
making the benchmark average 5331.2 and the ratio 0.9

 

[marcus]

I just checked as of noon 23 August and the index was 0.7,
which is what Ed Aboud predicted it would be for 1 September.
The notable thing, more than anything else, is the stability. The book doesn't seem to be going away at least not yet.

At the beginning of this month there was a shakeup in the fundamental physics popular market with the appearance of two new titles that seemed likely to appeal to readers. One of them, the Complete Idiot's Guide to String Theory (written by an editor of the Scientific American, no less) seems not to have caught on as yet. The other, not really a string book though I'm counting it as such because it has some stuff at the end, has been a great success: Susskind's book about the black hole controversy.

Despite the shakeup, Trouble sales ratio seems basically steady. It has been more volatile lately, I think, but as Gokul suggested I am going to do some averaging over a brief 7-day window and as far as I can tell (from occasional sampling) on average it hasn't changed much.

 

[marcus]

As of noon 24 August the ratio was 0.62

For the first time we will need the extra precision to distinguish who is closest, Mormonator (who said 0.5) or Ed Aboud (who said 0.7)

In this case, Ed would be the winner, because 0.62 is closer to 0.7.

What this tells me is that when I follow Gokul's advice and calculate an average
I will need to keep the un-rounded-off data---the unrounded ratio for several days, so that the average is not seriously contaminated with round-off error.

 

[marcus]

as of noon 26 August, the ratio inexplicably shot up to 1.1!

Trouble ranked 7546
and the five most popular stringies were
black 914
elegant 6468
fabric 6473
hyperspace 11,539
warped 15,504
for an average of 8179.6
making the ratio 8179.6/7546 = 1.1

*Mheslep* has registered the wildly optimistic forecast of 2.0!

That could actually happen as a brief one-day spike. But in response to a thoughtful suggestion (post #26) I have decided to make a SEVEN-DAY AVERAGE be the official 1 September target. It would be amazing if the average gets much above 0.7, I think. Averaging favors conservative middle-of-the-road forecasts. I say this with some regret, and apologies to Mheslep (who doesn't stand any more chance than a snowball in Hell, unfortunately.)

Well, now we are four who prophesy the future. Good luck to all!

 

[marcus]

Getting down to the wire

I start averaging as of 29 August, in two days. To me it looks pretty clear that Ed Aboud will be the winner. It will average out to around 0.7 (what he predicted). Today at noon the index was 0.74. I'm getting in the habit of recording two decimal places so the average will be more accurate (less roundoff error.)


1 September 6.4
1 October 6.5
1 November 5.2
1 December 2.4
1 January 1.5
1 February 1.3
1 March 0.4
1 April 0.6
1 May 1.0
1 June 1.0
1 July 0.5
1 August 0.4
...
...
23 August 0.7
24 August 0.62
...
26 August 1.1
27 August 0.74
...
1 September ??

 

[marcus]

Well tomorrow, 29 August, we begin to record noon readings that count in the 7-day average.
Today seems like the last time anyone could reasonably enter a forecast (the riskiest forecasts are the ones made a month or two in advance, as with Mo and Ed, but up til now there is still some uncertainty.)
Today at noon the index was 0.76

 

[marcus]

29 August 0.61
30 August
31 August

1 September

2 September
3 September
4 SeptemberAs suggested, the 1 September index will be determined by averaging the 7 noon readings on and around that date. Today 29 August is the first day contributing to the average and the salesrank ratio at noon was 0.61.

In case anyone is curious, the five currently most popular stringy books were black hole, parallel, fabric, elegant, and hyperspace. The ratio was 7259.8/11,930 = 0.61

Mo and Ed are the pessimists in the poll, and it's pretty clear now that the contest is between them. Mheslep and myself are the optimists and I regret to say I believe we are out of luck.