I don't know about the others, but I can see at least with Bastardi's Wager, they went with satellite data. That proved to be wise in retrospect. As regardless of which you use, lower troposphere satellite data has shown much less warning than the land/sea models used by the NOAA and the like. For them, 2015 was the third warmest year, and 2010 and champion 1998.
Except the land and ocean temperatures used by NOAA aren't modeled. Satellite data is modeled.
Satellites don't measure temperature. The measure microwaves. Temperatures for broad swaths of the atmosphere are derived by running the microwave radiances through a model to derive temperature. That's one reason why using satellite data as your primary temperature data source a bad choice. Not only does it not measure surface temperature, it also isn't very accurate and must be constantly adjusted.
Also, since they do measure broad swaths of atmosphere you're not going to get nearly as much of a warming signal as surface measurements. As GHG's increase, more heat is trapped closer to the surface, cooling the atmosphere above it. If you measure a swath, most of the warming gets averaged out. Without proper calibration against surface measurements, the difference between satellite and surface measurements becomes quite large. In fact, there was paper not too long ago that ripped apart Spencer's methodology as inherently flawed, showing how his method wasn't even close to actual observations.
To actually make wagers expecting cooling seems extreme. Why not simply bet that warming would come in far short of the predictions mainstream scientists were putting out at the time? Maybe they couldn't get anyone to bite on those terms, maybe they were just that cocky, or most likely, they just wanted the media from putting money up predicting cooling.
Whatever the reason, if they'd wagered on more sane "you're models show too much warming" terms, they could have made some good money.
Scientists aren't stupid. They wouldn't take that bet unless the period was climatologically relevant (30 years or so).
The problem is deniers often have little to know understanding of basic physics. If they had, they would have never made these bets. It was a free giveaway to the very scientists they so despised. It would be like someone coming up to you and betting you $10K that the moon is made out of cheese.
And no, the models don't show "too much warming". The models have under-predicted the amount of thermal energy being stored by the system. Or are you just referring to air temperatures (which is poor indicator of warming).