I’m curious: how do you decide where to submit your papers?

I’m perfectly willing to admit, my algorithm is roughly as follows:

  1. Try to get my coauthors to decide instead of me.
  2. Failing that, try to figure out a mathematician who is likely to edit some journals, and who would like the paper and know who should referee it.
  3. Find out what journals they edit (harder than it should be, though Google is OK for this purpose).
  4. If I’m happy with one of those journals, submit there.  Otherwise, back to step 1. or 2.

I’ve never heard a more sensible algorithm than this one, though I hope such a thing does exist.

39 thoughts on “Submissions

  1. I pick the journal first, then figure out what subject matter would be most likely to be accepted there. “Chaos, Solitons, and Fractals” is my next target.

    In this universe, though, once I’ve written a paper, I look through the references for the journal that appears the most, and usually submit there.

  2. I was just going to mention Allen’s second algorithm above, since that is the one he told me long ago. The only problem with it is if I followed it religiously, then all my papers would be submitted to Duke (I have submitted 5 papers there).

  3. Being much less senior than other people leaving comments, my algorithm is
    1) Choose prestigious journals. The manuscripts are swiftly rejected, and the reviewer comments are often strange.
    2) Submit to journals which have editors who actually care about the problem being addressed (Ben’s Step 2).
    Going straight to step 2 would seem sensible, but it’s worth buying the lottery ticket, I think, because having a paper in a more prestigious journal translates (at least in theory) to better job prospects, and makes no difference to readership because people who want to read the paper download it off ArXiv anyway.

  4. I’ve heard various people suggest Allen’s algorithm. Although as Joel suggests, this can lead to submitting all one’s papers to the same journal. This is, I’d guess off the top of my head, especially true if there’s a core bunch of papers that one tends to cite over and over again.

    Right now I’m doing basically what Andy suggests: try to get my advisor to decide instead of me. He’s not a coauthor, but he is familiar with my work. Of course my particular method only works if one *has* an advisor, and the portion of a typical career in which one has an advisor (therefore not having a PhD yet) but is submitting papers is hopefully short.

  5. I’ve had my PhD for a couple of years, and I still ask my advisor about where to submit papers. It’s not like your relationship with them ends upon graduation!

    Of course, one should also work on acquiring other mentors (this should start even before you graduate), and you should hit them up for advice as well…

  6. And if you can afford to be picky: Avoid journals where you have to use a web interface for everything.

  7. I somtimes find the rule-of-thumb “see one play one” useful. If a journal has accepted similar papers (of yours?) before there’s good odds they will do so again, though more generally I agree with the above statments of “ask my elders and betters” and “aim high”.

    You may, of course, have moral concerns too. If you’re principaled and wish to avoid Elsevere-esque journals then do so.

  8. Andy, that’s a good point. I just feel that it would be presumptuous for me to comment on what people do *after* they have a PhD, seeing how I don’t yet have one.

  9. I personally look for journals which publish articles in my area. I compare the articles recently published there with the one I want to submit, and if it looks like my paper is roughly of the same caliber and an appropriate subject matter I submit. I figure most people think more highly of their own work than others will, so this process will cause one to aim slightly higher than where you’ll end out publishing on average, so after at most say 2-3 submissions you’ll have found the “right” one to publish in.

    At one time I took a more idealistic approach, but I think you have to worry about career/reputation issues to some degree. I’ve found though that in any field there are only several “usual suspect” journals so it doesn’t take me very long to decide.

  10. Joel – perhaps I’m missing something obvious, but could you give a reason for this? I’m not so experienced with submissions, and I don’t understand why a web interface would be specifically bad.

  11. Yeah, I have to admit that I’m confused my the hatred for web submissions. I hate technology as much as the next guy, but at least they obviate the need to write the boilerplate “paper submission” email or letter.

  12. I’d certainly prefer a journal require web submission over requiring paper copies through snail mail. The former wouldn’t change my likelihood of choosing that journal, the latter might have if one of my coauthors hadn’t dealt with it. The one disadvantage I’ve noticed of web-submission over email is that with email if you cc coauthors the editor might reply all on the whole thread, which is convenient.

  13. I don’t think any journals require submission via snail mail. So the choice is between sending a quick email to an editor or going through some long web interface where you have upload a tex file, then upload separately an abstract, then enter all the authors names, and then wait for it turn it into a pdf and then “approve” the pdf.

    More seriously, another factor I consider when selecting a journal is how long the referee process will take. This is, of course, a bit hard to determine but I think that you can look at stats of the average time to acceptance/publication or just go by personal experience and/or hearsay.

  14. For the paper I’m about to submit, I asked people who know me and work in the same field which journals they recommend. For instance, my advisor suggested I talk to a former student of his who graduated shortly after I started grad school.

  15. I don’t mind a web interface, as it most likely makes it less likely for the journal to lose your submission. You don’t want to contact the journal after nine months to ask about the progress only to hear, “Oh, we don’t have a record of your submission.”

  16. Some web interfaces are less painful than the ones described by Joel (which are basically Elsevier’s and Springer’s, right?). The LMS web interface is fairly straightforward, for instance. No approving of PDFs or other such nonsense.

  17. I agree with Richard, having had precisely that experience, though luckily we asked after one month instead of nine.

    Also, I have a lot of trouble imagining a submission system onerous enough to compete with actually writing the paper. Though perhaps this is something about my psyche. Ever since I started primarily doing research, certain kinds of busywork have started seeming soothing, so a complicated submission procedure feels like a good cool-down from actually writing.

  18. Some journals (I can’t remember the specific examples) still claim to require snail mail submission, but if you ask an editor you will be told to submit by email.

    My only own experience with a lost submission was due to an editor’s changing email address, and my only own experience with a web form (for an independent electronic journal) was completely painless.

    I guess my overall point is that, with respect to these issues, Your Mileage May Vary.

  19. Some journals (I can’t remember the specific examples) still claim to require snail mail submission, but if you ask an editor you will be told to submit by email.

    Acta actually requires snail mail submission (to Sweden, no less). I have tried to submit by email, and was shot down.

  20. If you’re Acta you can get away with this.. but most journals would not want to discourage submissions by requiring international snail mail submission. I know I’ve decided to not submit to a journal because they wanted it in some annoying format.. this is in the same category and surely would cut down on their pool of papers.

  21. Kobe Journal of Mathematics also requires snail mail submission. They also require that the images be submitted separately, so as to physically “cut and paste” them into the text, I suppose.

  22. My usual is

    Dear Sir or Madam: I would like to submit my paper “Every uniformly bounded floozle has a hoodickey”, co-authored with J. Smith, for publication in the Duke Mathematics Journal. Thank you for your consideration.

    If a journal asks me for referee recommendations, I would also put them in the cover letter, but I don’t think Duke does that.

    Maybe there is a better way to do it, but this seems to have worked well enough so far.

  23. Duke wants a 300-word cover letter together with my paper.

    This sounds like you got some wires crossed; the 300 word limit is on the abstract length. I also support David’s cover letter. These things are 100% pro forma.

  24. Sorry – I only saw your replies now. I asked my advisor and he told me pretty much the same thing. I specifically addressed it to the editor I wanted to send it to, but otherwise it’s almost identical to David’s form.

  25. I’m surprised — especially given the “evil journal, good journal” tags on this post — that no one has mentioned the following: among the group of journals that would be reasonable homes for your paper on mathematical grounds, choose the one that doesn’t charge your library exorbitant subscription rates.

  26. It’s sometimes hard to know how much a journal charges. Is it a reasonable alternative to stay with university presses and avoid Elsevier and Springer?

  27. Is it a reasonable alternative to stay with university presses and avoid Elsevier and Springer?

    Yes, as a rule of thumb, university presses and learned societies are good while big commercial publishers are bad. (I’d also put AK Peters, a small commercial publisher, into the good category, but few others would join them there.)

    Besides the issue of cost, there is also the issue of copyright. I always insist on the ability to keep the final version of the paper in the arXiv, for example. Consdering that you are giving your paper to the publisher for free, there’s no sense in giving up more rights than you have to.

    Incidentally, you can find journal price comparisons at They also give the number of pages published for normalization purposes, although that’s not always a fair comparison (price per character helps correct for different page sizes and typesetting styles).

    Just to give you a sense for the range, Annals of Mathematics charges $0.13 per page, and JAMS charges $0.24 per page, while Functional Analysis and its Applications charges $7.12 per page. Last year 56 math journals cost $1.30 or more per page. With the possible exception of one journal translated from Russian, which may legitimately have higher costs, this is ridiculous (even if, hypothetically, their pages are twice as big as those in the Annals). Aside from the Russian journal and one journal published by “IOS Press” (which I’ve never heard of), the other 54 are all published by the usual big commercial publishers (Elsevier, Springer, Wiley, Taylor & Francis, etc.).

  28. It’s sometimes hard to know how much a journal charges.

    Actually, it’s not. Just bookmark this page.

    Is it a reasonable alternative to stay with university presses…

    It’s reasonable, especially since I think there are good reasons to support non-profit journals over for-profit ones, even with comparable costs. It is important to keep in mind that there is still wild variation amongst non-profit journals for costs, for reasons that remain unclear. Oxford and Cambridge University Presses have journals that cost enough to probably be incompatible with being a Banff Protocol signatory (for example, the most expensive journal I have published in is an Oxford journal, even though I have papers in Springer and Elsevier journals).

  29. For anyone interested in building a personal database of information on journals, I’ve put together some instructions on how to do it here. (It has to be instructions rather than providing you with my database due to copyright issues).

  30. I’d like to thank David Speyer for his sample cover letter (comment #29 above), because I just used it (of course with the title of my own paper, and my own name!) to submit a paper.

  31. The requests thread has grown too long by now so I guess I’d better post the request here. Could anyone please make a post gathering the information about the math journals’ rejection rates? Thanks in advance!

Comments are closed.