[sterlingice]

Sent.

Far be it for me to question the undisputed IWS master, but I wanted to add some thoughts as to why it took me so long. Also, I know it's not cool to table talk in IWS threads for fear of influencing the answers so I'll try to speak in generalities.

I felt the "name the pick number" questions were really tough. I just gave up and went and pulled up a random free mock draft and tried to go off of that. I think outside of the top 10 picks or maybe the top couple QBs, unless you've been actively wargaming picks, it's just a shot in the dark. And even then, there's a ton of variance.

I had a theory a while back about IWS questions about how most are either eliminators or splitters. Eliminators are the ones that have one real answer that the majority picks, but eliminates the rest - if you don't get this question, you're out of the running. However, it can't be a total slam dunk or it serves no purpose: if, say, 19 people get it right and only 1 wrong - it doesn't eliminate much. A good eliminator still shrugs off maybe 1/4 of the entries.

Meanwhile, a splitter is one with no real consensus but a little of you need to get one of the top couple of answers on each of the splitters to win.

In a perfect IWS, I think there's a balance with at least two of each and two that could go either way. You want the eliminators to make it feel like the winners and losers "earned" their spots while the splitters should keep you from having too many ties.

To roll this back here, eliminators are usually ones that people can come up with off the top of their head and splitters are ones they have to look up and do extra work. The more splitters, the less likely people are to participate.

The "which pick" questions are almost all splitters and I suspect there are going to be a of 1s and maybe 2s with those for scores. For instance, making the Matt Araiza question a pick number makes it a complete dart throw (and definitely a splitter). Whereas maybe asking what round would have made it an eliminator or a hybrid splitter where you get a clustering of answers in 3 or 4 buckets. Same with the running back pick- maybe have people pick a bucket of 10 it will go in and not an exact pick (1-10, 11-20, 21-30, etc).

SI