Last Monday night saw the premiere of A Cappella BEATdown — LIVE!* It was by far the biggest thing I’ve produced for WILL since my late and unlamented weekly movie review series Critics’ Choice went off the air in 2002. And a typical Critics’ episode had nothing on a live, three-hour music competition and pledge-o-tainment show featuring nearly 100 young singers!
On the whole, it went very well. There were no major technical kerfuffles**, all of the groups wound up where they needed to be when they needed to be there, and the music sounded great. Really, we have a lot for which to be proud.
However, me being me, it’s hard not to overemphasize the things that went wrong. We had the wrong phone number up for the first couple of songs. The ringing phones proved to be too loud to peaceably coexist with the music. Midway through the production, we mysteriously went from being three minutes ahead to ten minutes behind my carefully-planned timetable. And, as we had set our online poll to automatically shut off a few minutes after the final group originally had been scheduled to perform, I frantically tried to reach our IT staff to extend the deadline.
None of that compared, however, to my sudden realization that the online poll had been utterly compromised. I’d been refreshing the results on my laptop, and with about a half hour to go, I found that one of the groups suddenly shot from fifty or so votes to more than a thousand! The website began to bog down as the total kept climbing: 1,500…1,800…2,100! It was clear that the we were being spammed.
Now, we had set up the poll to discourage multiple votes, but had been advised not to limit it to one per IP address as doing so could disallow some legitimate participation. Instead, we had a cookie-based restriction. I knew that there’d be ways around it–for example, one could cast a vote in each different browser–but I thought that the relatively short time window and low stakes of the competition might be enough to keep someone from trying to be tricky.
Ding dong, I was wrong.
With only minutes left until the announcement of the winner, that left me with a dilemma. Should I honor the obvious fraud? There was no time to sift through the votes and remove the duplicates. And as the online poll displayed the current results after a ballot was registered, the group in question could see that they had “won” that vote.
We did have one out, which was that we had deliberately kept the online poll separate from the phone-in poll. We knew that it would be faster to make a web vote than to call the pledge line. (Indeed, while we logged 766 phone calls that night, we had 5,003 online ballots.) And in the case that the “winners” of the two polls disagreed, our judges would break the tie.
Unfortunately, my wife Vicky–whom I’d drafted as the queen of the phone poll–was having a hell of a time tabulating the results. Meanwhile, our host began to vamp as everyone waited for the verdict.
I conferred with the judges, who agreed that the leader of the phone vote (a very talented Hindi group named Chai-Town) should be the overall champion. I scurried off to tell the production staff to bring the singers into the studio…but neglected to give the nod to our host, who kept tap-dancing away.
The eventual announcement of the winner was awkward and quite possibly a bit confusing for our audience as Chai-Town seemingly came out of nowhere.
In the days that followed, we were able to extract the online poll log and pull out the duplicate votes. We found that while the more than 2,100 bogus ballots for one group was the most egregious example, most of the teams benefited from the spamming of the server. Two other groups received in excess of 500 duplicate votes.***
Once the duplicates were deleted, we were down to 983 ballots. Happily, we discovered that Chai-Town won the web poll as well as the phone poll and the judges’ vote! Potential crisis averted!
The muddled conclusion of the show and the middling results–only $3,395 raised during three hours of airtime–left me unable for a time to consider what I and my coworkers had managed to pull off. Yet, I have to admit to myself that it was one of our more ambitious productions. We had our first online poll, a live web stream with somewhere between 300-500 distinct viewers, plus Facebook updates leading up to (and even during) the program. Everyone seemed to have a good time, and most of the folks I talked to said that they really enjoyed the show. Most important to me, we got nearly 100 young people into our studio. Presumably even more watched and participated. I’d wager that for many of them, it was their first exposure to WILL-TV in a long time.
It may be some time before I can bring myself to watch my DVD copy–especially those last few minutes of tap-dancing–but in the end I expect that I’ll feel pretty good about the experience and even ready to try it again, WILL willing.
*The exclamation point was part of the title so that you knew how excited to be about it.
**An actual industry term. Look it up.
***Just to be clear, we’re not suggesting that the contestants themselves were cheating. Several someones appeared to be doing so on their behalf, and quite likely without their knowledge.