I am new here, so if this is not appropriate place for such question, the tags are wrong, or if this is a duplicate (I have checked, but found nothing, still absence of counterexample is not a proof), just let me know.

As a new member of SE society I have observed that users with high (or should I say extremely high?) reputation have greater average of votes per answer than the rest. This is great, but I am curious what is the secret? It is sure obvious if the problem is complex or the solution is particularly brilliant (and high reputation indirectly means high skill and keen mind). But those are not the only cases. Are there some do-s and don't-s that might help (I have read the FAQ)?

This is especially important to me since I am a grad student and I teach in "real". In a class I am doing rather fine (some tricks may also apply here so I am eager to hear them), but with written text I am not so good (to be honest, I may call that one of many various reasons I finally subscribed to SE), so probably there will be similar problems with any article I write.

I am aware that there are many (too many?) books on teaching and writing (i.e. I kindly ask you to mention at most one per answer), so what I am asking for is your personal experience. I will be very grateful if you could share some of your secrets or if you have any advices.

To not be one sided I can offer an advice from Stephen King (actually his book on writing) that always worked particularly well for me: "2nd Draft = 1st draft – 10%".

• Generally the highest voted answer is the quickest (reasonable) answer. Very frequently later better answers get few if any votes. It's truly unfortunate that this is how the site works, but there is little that can be done to change this without major redesign of the software platform (which we have no control over). In any case, one shouldn't be misled into thinking that the highest-voted answer is the best pedagogically. Often the opposite is true. Also deeper answers tend to get fewer votes. Votes and rep are not good measures of anything. – Bill Dubuque 12 mins ago – Bill Dubuque Mar 8 '12 at 21:58
• @BillDubuque Do you claim that you earned your 50k by good reflexes, fast typing and superficial answers? Sorry, I do not believe it! :-P – dtldarek Mar 8 '12 at 22:01
• Alas, if you peruse the answers of many experienced mathematicians, you will discover that many of their best answers received few votes, and many of their highest voted answers are, indeed, "superficial". – Bill Dubuque Mar 8 '12 at 22:24
• As someone who got $98$ upvotes for explaining why the probability that the second roll of a die yields a higher number than the first is $2.5/6$, I find it hard to argue with the claim that the highest voted answers tend to be superficial, or at least, extremely elementary and not requiring any mathematics I learned in my adult life. This is kind of in the nature of things, though, and I have long since gotten used to it. – Pete L. Clark Mar 9 '12 at 0:27
• @Bill is known in some circles at the FGITW :) – The Chaz 2.0 Mar 9 '12 at 2:36

My conjecture:

More experienced mathematician => knows more things and knows certain things better => chooses to only answer questions in which he or she can quickly provide a reasonably worded answer.

Basically, it is an issue of self-selection: I won't bother to answer questions where I cannot quickly provide a good answer. (There are times when I even abort writing an answer for a question in a subject for which I am an expert because I realized after the second paragraph that to do the question justice requires time that I just didn't have at the moment.) So when I do answer questions the answers often end up being both (a) one of the first ones and (b) at least decent in quality.

As a corollary to the above, and since you mentioned that you are a graduate student: the content of a great talk or a great paper often is only a compact subset of what you know. At the boundary of your knowledge, there are a lot of things which, if you look back at it a couple months or a couple of years from now, after you've thought about it more and really pushed the boundary further, you will be able to see in a simpler and more concise way. (This is also why review articles and textbooks exist: on hindsight a lot of things in journal articles can be made simpler just by a more modern language/point of view.)

This is actually a great thing about teaching as a graduate student: it exposes potential weaknesses (if you have trouble explaining something, then you don't really understand it completely) which you would've thought should be extremely simple to you (calculus, linear algebra, basic analysis). I know that it wasn't until I taught undergraduate linear algebra and was preparing lecture notes that a lot of things that I picked up in geometry and functional analysis clicked into place and allowed me to really understand some things.

• +100 on that last paragraph. I cannot stress how much writing answers on MSE has helped my teaching, and vice versa. – Asaf Karagila Mar 9 '12 at 10:54
• @WillieWong Thank you for you answer. – dtldarek Mar 9 '12 at 12:50
• @AsafKaragila Could you explain how did it help you and possibly how your style of writing/teaching has changed since you started to write on MSE? If you have time, I would prefer longer answer than short comment (it won't go unappreciated). – dtldarek Mar 9 '12 at 12:52
• @dtldarek: I don't really have the time, so I'll just leave this comment here, consider the time you spend on an answer. On a good answer. on an answer that explains something in a deep way. How long would it take you to formulate it? If it's the first time you've explained this - could be a whole day. If it's the second time, you'll take a bit less. When writing similar answers for the tenth time, you can do it in half an hour. Similarly when answering students. When you have both the students and this site, you can improve twice as fast. – Asaf Karagila Mar 10 '12 at 14:50
• @AsafKaragila Thank you for your answer, especially if you don't have much time! – dtldarek Mar 10 '12 at 19:45

It is not clear to me whether you are asking, "How can I get more votes on m.se?" or "How can I improve my writing?" These are two very different questions. But I suppose that the advice, "practice! practice! practice!" applies to both questions.

• My question simplified can be expressed as "Is there any particular reason you get so many votes on average per post on SE?" or maybe even better "What type of writing is the most understandable (appealing?) to people?". – dtldarek Mar 8 '12 at 23:41