EA Forum Prize: Winners for February 2019

CEA is pleased to an­nounce the win­ners of the Fe­bru­ary 2019 EA Fo­rum Prize!

In first place (for a prize of $999): “Ev­i­dence on good fore­cast­ing prac­tices from the Good Judg­ment Pro­ject”, by koko­ta­jlod.

In sec­ond place (for a prize of $500): “Small an­i­mals have enor­mous brains for their size”, by eu­kary­ote.

In third place (for a prize of $250): “Will com­pa­nies meet their an­i­mal welfare com­mit­ments?”, by saulius.

We also awarded prizes in Novem­ber, De­cem­ber, and Jan­uary.

What is the EA Fo­rum Prize?

Cer­tain posts ex­em­plify the kind of con­tent we most want to see on the EA Fo­rum. They are well-re­searched and well-or­ga­nized; they care about in­form­ing read­ers, not just per­suad­ing them.

The Prize is an in­cen­tive to cre­ate posts like this. But more im­por­tantly, we see it as an op­por­tu­nity to show­case ex­cel­lent con­tent as an ex­am­ple and in­spira­tion to the Fo­rum’s users.

About the win­ning posts

Ev­i­dence on good fore­cast­ing prac­tices from the Good Judg­ment Pro­ject” is a thor­ough, well-or­ga­nized sum­mary of fore­cast­ing — a topic of­ten dis­cussed on the Fo­rum, but rarely with this amount of data.

We may know that pre­dic­tion mar­kets are “use­ful”, but the au­thor goes far be­yond that, ex­plain­ing how well differ­ent types of mar­kets (and non-mar­ket mechanisms) have performed in pre­dic­tion tour­na­ments, and which char­ac­ter­is­tics the best fore­cast­ers tend to have. This re­search could be use­ful to any num­ber of fu­ture fore­cast­ing pro­jects in the com­mu­nity.

Ad­di­tion­ally, the au­thor:

  • Uses num­bered head­ers to sep­a­rate sec­tions.

  • In­cludes hy­per­linked foot­notes for all cita­tions.

  • Notes cases where in­for­ma­tion from origi­nal sources is miss­ing or un­cer­tain, giv­ing read­ers ideas for ways to con­tribute to his re­search. (For ex­am­ple, I’d love to learn more about Tet­lock’s “per­pet­ual beta” con­cept, if any­one cares to go and find it.)

Over­all, this is a re­mark­able post, and I hope that other Fo­rum users cre­ate similarly ex­cel­lent sum­maries of im­por­tant con­cepts.

Small an­i­mals have enor­mous brains for their size” makes a sin­gle, sim­ple point (you can see it in the ti­tle), but does so with un­usual el­e­gance.

I still re­mem­ber the core simile — “you have as many neu­rons as a half-full bucket of ants” — many weeks af­ter I first read the ar­ti­cle, and ex­pect to re­mem­ber it for years to come, thanks to the origi­nal art which en­livens the piece. Illus­tra­tions aren’t es­sen­tial to Fo­rum posts, but mak­ing good ideas mem­o­rable, how­ever you choose to do it, am­plifies their im­pact.

Ad­di­tion­ally, the au­thor:

  • Recom­mends fur­ther read­ing for any­one who found the ar­ti­cle in­ter­est­ing (this is sur­pris­ingly rare for EA Fo­rum posts, de­spite the vast liter­a­ture that in­forms many of our ideas).

  • Doesn’t over­state her point; in­stead, we get facts about neu­rons, plus a list of ways in which these facts could in­ter­act with cer­tain be­liefs to pro­duce other be­liefs, with­out ad­vo­cacy for any of those be­liefs.

    • There’s noth­ing wrong with ad­vo­cat­ing be­liefs, of course, but there can be ma­jor benefits to sep­a­rat­ing “fact posts” from “be­lief posts”. For ex­am­ple, a fact post is more likely to be cited by au­thors with a range of be­liefs, mak­ing ev­ery­one’s be­lief posts more ev­i­dence-based in the pro­cess.

Will com­pa­nies meet their an­i­mal welfare com­mit­ments?”offers cru­cial con­text on one of the most pop­u­lar causes in EA: an­i­mal-ad­vo­cacy cam­paigns tar­get­ing cor­po­ra­tions.

If com­pa­nies don’t ac­tu­ally live up to their promises, we haven’t made an im­pact. The au­thor pulls to­gether dozens of differ­ent sources from in­side and out­side of the EA com­mu­nity to show that… well, these promises may not be as im­pact­ful as they first seemed. But he doesn’t just ex­plain the is­sue; he also notes the high level of un­cer­tainty around par­tic­u­lar facts and figures (pro­vid­ing bet­ter in­for­ma­tion even at the risk of un­der­cut­ting his “point”) and sug­gests ways to im­prove the situ­a­tion.

Ad­di­tion­ally, the au­thor:

  • Uses our built-in header sys­tem to sep­a­rate sec­tions (I’m re­peat­ing my­self here, be­cause this is a re­ally use­ful fea­ture and I strongly en­courage au­thors to use it for any­thing longer than a page or so).

  • Pro­poses im­prove­ments that an­i­mal char­i­ties could make with­out harshly crit­i­ciz­ing those char­i­ties (dis­t­in­guish­ing be­tween “things could be bet­ter” and “things are ac­tively bad” is a good habit).

  • Points out the ways in which his find­ings might af­fect our cost-effec­tive­ness es­ti­mates around an­i­mal ad­vo­cacy. Ex­plain­ing a cru­cial con­sid­er­a­tion is good; es­ti­mat­ing its im­pact makes the ex­pla­na­tion even bet­ter.

The vot­ing process

All posts pub­lished in the month of Fe­bru­ary qual­ified for vot­ing, save for those writ­ten by CEA staff and Prize judges.

Prizes were cho­sen by six peo­ple:

Vot­ers re­cused them­selves from vot­ing on posts writ­ten by their col­leagues. Other­wise, they used their own in­di­vi­d­ual crite­ria for choos­ing posts, though they broadly agree with the goals out­lined above.

Win­ners were cho­sen by an ini­tial round of ap­proval vot­ing, fol­lowed by a runoff vote to re­solve ties.

The fu­ture of the Prize

After re­view­ing feed­back we’ve re­ceived about the Prize, we’ve de­cided to con­tinue giv­ing it out for an­other six months (Fe­bru­ary through July) be­fore run­ning a sec­ond round of re­view. We don’t have any cur­rent plans to change the for­mat, but we won’t rule out po­ten­tial changes in fu­ture months.

If you have thoughts on how the Prize has changed the way you read or write on the Fo­rum, or about ways we should con­sider chang­ing the cur­rent for­mat, please let us know in the com­ments or con­tact Aaron Gertler.