Follow This Blog: RSS feed
Heavy Medal: A Mock Newbery Blog
Inside Heavy Medal

How does a book Win? Part 3: Balloting

496px-your_vote_counts_badgeSo, we’ve nominated our books and we’ve discussed them and discussed them and discussed them. Now, it’s time to vote!  But how does that voting work?  And how does the Committee decide what books are honor books?

Let’s start with our Newbery winner.  In order to be selected as the Newbery winner, a book must have at least eight first place votes (a majority) and must have an eight point lead over all other books.  (In our mock discussion, we will base these numbers on how many people show up and are part of the discussion. For the real committee of 15, this is a majority, so for our discussion we will use a small majority as the number required, as well.)

screen-shot-2017-01-08-at-10-20-19-pmOnce all books have been discussed, the members of the committee each write down their 1st, 2nd, and 3rd choice books on a ballot.  This ballot is tallied.  Each 1st place vote = 4 points.  A 2nd place vote = 3 points.  A 3rd place vote = 2 points.  If this first ballot produces a winner, then voila!  We have a Newbery medal winning title!  If not, then re-balloting is necessary.

Before another vote can be taken, though, the committee MUST have more discussion.  During this discussion, by consensus the committee may choose to remove books from the table.  They will likely remove books that had no votes, and with discussion it may make sense to also remove books with few votes.  If those books are unlikely to suddenly rise to top, it is helpful to take those few votes they got and disperse them among other titles that are more likely to win.

12_angry_men_5We never know what goes on behind the doors at the Newbery table, though.  A 12 Angry Men situation could certainly happen with one or two members swaying the committee via discussion of titles.

Re-balloting will happen as many times as need be to get the 8 point spread and the 8 winning votes that produces a winner.  HOORAY!  A WINNER!

This is what the Manual tells us about re-balloting:

The committee may not proceed to another ballot without a second round of book discussion. At this point, certain choices present themselves, and certain procedures apply:

  • By consensus the committee may choose to withdraw from the discussion list all titles that receive no votes on the first ballot.
  • By consensus the committee may choose to withdraw additional titles that received minimal support on the first ballot.
  • Once withdrawn from the discussion list, a book is permanently eliminated from consideration for the award.
  • Once a second round is complete, the committee proceeds to a second ballot.
  • On a second ballot (and, if necessary, all subsequent ballots), votes are tabulated by the tellers who use the same point system and formula as in the first round to determine a winner.
  • If after a second ballot, there is still no winner, the committee is required to re-open discussion and then re-ballot, alternating between discussion and re-balloting until a winner is selected.

14631799906_04eb1048a6_bAt that point, the committee decides on honor books.  The committee can name any number of honor books they choose, including none.  They can select those honor books from the final ballot used to select the winner, or they can ballot again for honor books.  Here is some of what the Manual says about honor books:

These shall be books that are also truly distinguished.

…as well as…

The committee may name as many or as few as it chooses, or none, keeping in mind that the books should be truly distinguished, not merely general contenders. 

We are getting so close to knowing the winner – our online winner, our in-person winner, and the REAL winner!





Sharon McKellar About Sharon McKellar

Sharon McKellar is the Supervising Librarian for Teen Services at the Oakland Public Library in California. She has served on the Rainbow List Committee, the Notable Children’s Recordings Committee, The Arbuthnot Honor Lecture Committee, and the 2015 Caldecott Committee. You can reach her at


  1. What is the harm in making this public, following the voting? It would be fascinating, to me, to see some real ballots and data and see what paths some winners took to win.

    • Sharon McKellar Sharon McKellar says:

      There has been some discussion around this recently, and it’s an interesting topic to me. To some extent, I understand the desire for everything to remain enshrouded in secrecy, as it makes sure committee members feel safe speaking and voting their minds, and also it ensures that the winner is just seen as “the most distinguished” without there being a lot of, “yeah, but…. it took 8 hours and blah blah blah” making the award feel less strong.

      On the other hand, there is a lot of historical interest in what goes on in that room, and for good reason.

      • Jonathan Hunt Jonathan Hunt says:

        SLJ did an up for discussion piece in the spring. I think the issue is whether there should be increased access after a given period of years for the purposes of research and such.

        For a couple years in the late 70s, SLJ published the nomination list. Not sure why they stopped the practice, but it would probably be a welcome one today.

      • See, I wouldn’t necessarily need to see discussion notes or anything, just a peek at ballots would be enlightening, to see what was discussed. I would imagine there have been years that would surprise us. Meaning, titles we’ve devoted a lot of time to discussing haven’t found spots on ballots. Or vice versa, titles we have devoted a lot of time to discussing have immediately cemented themselves as front runners. (I’m sure that is more likely.)

      • Jonathan Hunt Jonathan Hunt says:

        Yeah, I don’t think that’s ever going to happen. I think the Newbery mystique is part of its appeal. I also think that allowing us to effectively be flies on the wall doesn’t serve the books and their creators very well. However, I would like to see a broader range of books (and a slightly higher number) bask in Newbery glory. Somehow.

      • I agree that the mystery shrouding the discussions adds to the appeal!

  2. Leonard Kim says:

    Are all aspects of the honor book selection procedure (whether to have honors, whether to re-ballot, how many honor books) simple majority vote? The Manual says the committee votes on these matters but is not specific. In particular, it’s not clear how you’d vote for how many honor books. Suppose 6 people felt there should be 1 honor book and 9 people felt there should be many honor books, but in a straight vote they split with 5 votes to have 4 honor books and 4 votes to have 5 honor books. Does the 1 honor book faction win?

    • Sharon McKellar Sharon McKellar says:

      That’s a good question, and there isn’t a straight answer for it, because, as you point out, the manual doesn’t tell us. That means each committee decides not only how many honor books, whether or not to re-ballot etc, but also how to decide on these things. Some committees may agree together that majority rules, and others may decide they need to have a stronger mandate from their members. So, in the example you bring up, hopefully the committee would discuss ahead of time how they will decide in a case like that and if not, they would talk and talk until they all agreed. So, in any case, it is a consensus in that the committee has agreed on how to decide, and then decides that way.

      I do think, generally, it’s not about deciding how many honor books, as it is deciding which of the non-winners are “truly distinguished” and looking at those books. In some cases it may be clear that, for example, everyone agrees 6 of them are and then the next 3 are clearly grouped together as equally distinguished, and the group needs to figure out if they belong in the honors or not.

      Nine honor books! Now that would be fun!

      • Jonathan Hunt Jonathan Hunt says:

        Yes, it’s frustratingly vague and each committee works that out. I do think that a straight majority is probably a good barometer, however. So, as a general rule, I might infer that honor books had approximately 8 people, give or take, supporting them on the final ballot. That’s probably not an iron-clad inference, but it’s one that we’ve used in our in-person mock Newberys.

    • It’s always greatly informed by the ballot results that they’re looking at – either their winning ballot or their honor ballot. So in the case you described, we could infer that the voting has left one clearly distinguished choice at the top of the results and then a gap of some significance before the next “many”. In that case, I would guess that perhaps it is the winning ballot, and that the one clear honor choice was amassing consensus alongside the winner. So many of those who had been swayed from their first choice as consensus was building might advocate for a separate honor ballot and spread their support back out. The weighted results are quite helpful in establishing the sorts of gaps that might seem significant to a majority of the committee. Even the most ardent advocate for having lots of honor books might find it difficult to argue for the inclusion of titles that fall after a 10 point gap.

      It would certainly be interesting to know, historically, how these decisions have been made – particularly in years when the number of honors deviates from the expected. Was there a particularly persuasive chair or committee member that convinced the committee to stick with only one honor? Did support spread so widely on the honors ballot that they had to include all those titles or none, or did they choose to include the first two clusters of titles, jumping a gap that a different committee might have chosen as their cutoff? It’s fascinating to speculate.

      • I don’t think the balloting would tell you much. The real happening is the discussion and there is no record of it. I personally think the secrecy of the ballot should be preserved. All the speculation fun is fun. Do we really want to lose it?

  3. I would love to have someone write on the plusses and minuses, no pun intended, of the arithmetic of the Newbery and Caldecott voting procedures. Any takers?

    • Jonathan Hunt Jonathan Hunt says:

      Meaning write a blog post? Or a magazine article? Hmmm. Thinking, thinking . . .

    • Eric Carpenter says:

      The voting procedures are a modified Borda Count (using 4,3,2 instead of 3,2,1 points and with an additional criteria of a majority of first place votes). I wonder if the horn book could find a math/game theory grad student at one of those Cambridge universities to do a deep mathematical look at the voting procedures and offer a few optimum strategies.

      • Jonathan Hunt Jonathan Hunt says:

        I second this idea!

      • Leonard Kim says:

        As Eric mentioned, The Newbery system is based on a Borda Count. In addition to the first place majority (and 8-point gap), another difference from a strict Borda count is that committee members do not rank all candidates, only three.

        I think these features of the Newbery system do a good job reducing some problems with a strict Borda count.

        1) the temptation not to vote your true preference and “bury” your perceived rivals. That is, if I had to rank all 100 or so nominees, if I really wanted WHEN GREEN BECOMES TOMATOES to beat GHOST, I might be tempted to vote GHOST last (as in 100th place) instead of vote honestly (where it might be say 15-20).

        2) the “mediocre” winner — a strict Borda count winner is supposed to be the points winner alone. Suppose we had 8 ballots that looked like this #1 GHOST #2 WHEN GREEN BECOMES TOMATOES; and we had 6 ballots that had #2 WHEN GREEN BECOMES TOMATOES and #3 GHOST (and something else for #1) and we had just one person leave GHOST off the ballot but ranked WHEN GREEN BECOMES TOMATOES #2. On a points basis, that gives WGBT 45 points and GHOST 44 points. So WGBT wins the Borda count even though *nobody* ranked it #1 and even though more people ranked GHOST ahead of WGBT then vice versa (8 vs 7), A strict Borda count allows results like this that might seem questionable, but the Newbery additions solve this problem by requiring re-balloting (where presumably, the GHOST faction would drop WGBT to allow the required margin while preserving the first place majority.)

        4) The polarizing weak majority: let’s say 8 people love SAMURAI RISING and vote it #1. 7 people hate hate hate it and leave it off the ballot. The points gap rule allows some safeguard from a technical majority winner that a significant minority is unhappy with. As long as those 7 people aren’t too divided and can come up with an alternative that gets at least 4 first place votes (say 4 1st place and 3 2nd place, or 5 1st place, 1 2nd, 1 3rd) then SAMURAI RISING can’t be declared winner yet and the discussion continues.

        So I like the Newbery system a lot. The only downside I see is that it may require many rounds of voting. I think we are all OK with that.

        For the casual reader, William Poundstone has a great book about the quirks of voting called Gaming the Vote that I highly recommend!

      • Leonard Kim says:

        So I suppose one “weakness” is that if you are trying to vote “against” a book you don’t want to win, you and like-minded readers need to be agreed on the specific, alternative candidate to support. As can happen in politics, if you vote with your heart, you may end up voting for a third-party spoiler and end up with a least favorite candidate instead of an acceptable compromise. i would presume the Newbery Committee can only discuss the books and that they cannot discuss how they plan to vote, so the sort of negotiation needed to really protect against this can’t happen. Nevertheless, voting for a spoiler is a hazard of any voting system, and the Newbery system overall does a good job against that (e.g., you need 8 1st place votes, not just the most 1st place votes, which really would be vulnerable to spoilers.)

    • Sharon McKellar Sharon McKellar says:

      Get Nate Silver on the case!

  4. Leonard Kim says:

    In terms of strategy, I would suggest one should only vote for books you think deserving of the Medal. If you think something is “just an honor book” I would leave it off the ballot completely. In this sense, the Newbery system is less like a Borda Count than another system called “approval” voting. If you believe 1 book is miles ahead of the rest, I’d use my 2-3 vote on something like an easy reader that you know may have little effect on the final result. Even if I like GHOST and would be fine if it won something, if I truly felt WHEN GREEN BECOMES TOMATOES should win the Medal, I should leave GHOST off the ballot to maximize the points distance. Any book you vote for, even with your 3rd place vote, helps it to some extent towards the Medal. After the Medal winner is chosen, you can always vote to re-ballot for honors and choose the rivals then.

Speak Your Mind