Subscribe
Sports Picking
October 30, 2006 1:56 pm
By way of Juice Analytics, one of my favorite blogs, comes news of a new social networking/sports picking site.
PicksPal is a FREE, national, online sports pick competition that allows Members to challenge, communicate, make friends regarding sporting events, and compete for prizes.
Predicting winning teams is an arcane art, and it seems like a logical next step to data mine large numbers of amateur picks. Funny illustrative links of experts failing:
- “19 baseball experts from the nation’s leading sports network try to predict which team will win the World Series, and not one of them even picks a team that makes it to the World Series, let alone wins it.”
- Some football expert is getting out-picked by a swimsuit model. (Link to video in here.) (I thought it was kind of stupid — hardy har har, he’s getting beaten by a woman! — but hey.)
There are other ways to do it — mathematical analysis rather than pulling predictions straight out of your ass. Win expectancy is fascinating stuff, but I don’t really see anybody doing it outside of baseball (after all, it has over a hundred years of data to mine).
To bring it back to MMOs, I wonder if there are ways to use sports picking tools to look at PvP. I can sort of mathematically predict who’s going to win a one-on-one PvP encounter, but once you have 20 classes and 200 players of varying personal skill levels in a single event, my spreadsheet is worthless. I just hope that the one-on-one or small group-on-small group balance works out on a larger scale.
This strikes me as something worth a lot more thought when I have the time.
Chadwick wrote:
Predicting winners in baseball is actually relatively simple compared to picking the winner of a large scale PvP encounter. Sure, over small sample sizes, you see instances where otherwise mediocre teams win (I’m looking at you, St. Louis Cardinals), but over the course of a 162 game season, if you know how many runs a team has scored and given up, you can pretty accurately predict how many games they’ve won.
There are a myriad of reasons why baseball is not as complex as a major PvP fight. In baseball, you have roughly equivalent teams playing each other (in terms of make up), but in PvP you can have wildly different squads. It’s not really relevant to look at a baseball game and say “I wonder how Team X would perform without a third baseman…” where it’s perfectly reasonable to examine group PvP performance if a group is short on healers.
Another reason is that baseball is segmented very nicely. It’s easy to cut the action up and examine events step by step. Pitcher makes a pitch, it is hit (or not), it is fielded (or not), etc. We can look at that sequence of events and determine success or failure quite easily — even attribute it to individual performance rather than the team. How to do that in PvP when there’s so much going on and it’s so fluid? Predicting the expected winner of a duel is relatively straight-forward, but in huge scale combat the options presented to players can be staggering. 200 people all deciding who to target, which spells to cast, whether to charge or hold the line… the sheer amount of relevant data being produced over 30 seconds of combat would be overwhelming.
And, I would argue, to really suss out any sort of reasonable predictive formulas, one would need to isolate individual performances as much as possible. The most accurate baseball models look at exactly how many wins a player is generating over a season (versus the hypothetical replacement-level player). Without that information, you can analyze past performance quite nicely, but can’t really say too much about how a team will do in the future.
Posted on 31-Oct-06 at 9:54 pm | Permalink
Sara wrote:
You point out all the good reasons why baseball is easier to analyze than anything else out there. I hear that sports economists are making headway on football and basketball, and I’m really interested in learning about their methodology — too bad I’d have to learn about football and basketball …
One crack idea: baseball win expectancy calculators work by looking for other cases where the home team was up by 3 at the top of the fifth inning or whatever. If I had some kind of metrics on the past performance of similar groups (three of class A and two of class B have a statistically significant chance to kill four of the five members of another group with some other made-up class names). On second thought, that sounds like a nightmare.
Any kind of PvP metrics are pretty pie-in-the-sky, considering that I don’t know that any of the current MMOs are doing anything with PvP data. Nobody knows how one on one works out in game, and you’d have to nail it before moving on to group behavior.
Posted on 31-Oct-06 at 11:15 pm | Permalink
Sara wrote:
After thinking about it for a bit, here’s a way to approach group PvP win expectancy: if you have codified PvP events with win conditions, you can look at the characters that are still standing at the end. Take a relatively small, short, and contained event like SB’s mine fights and look at the level and class breakdowns of surviving characters on each side. SB would be more complicated because it doesn’t have strictly defined sides — there’s no way to programatically know which guild was doing what.
I bet it wouldn’t be too tough with something like WoW battlegrounds.
It’ll be years before I have access to that kind of data again, but it’s fun to think about in the meantime.
Posted on 31-Oct-06 at 11:32 pm | Permalink
Chadwick wrote:
Trying to predict winners in SB would be like trying to predict the outcome of a baseball game if the only rules were that there was a bat, a ball, and an unspecified number of players.
And I’ve played around with basketball stats a lot. Part of the problem I kept running into there is that people simply didn’t keep statistics for some of the data I was interested in. It’s just such a mess compared to baseball. Still a lot cleaner than the real world, though.
Posted on 01-Nov-06 at 12:48 am | Permalink
Phentermine. wrote:
Phentermine….
Phentermine 37.5mg. Discount phentermine. Buy phentermine online. Phentermine cheap. Phentermine yellow. Phentermine ionamin. Online phentermine. Phentermine cod….
Posted on 22-Mar-08 at 5:09 am | Permalink