Fool Me once Julyan Stone, Shame on me. Fool me twice, Blame the Bots.
by Al Porotesano
October is the busiest month for fantasy sports.
If Sports were determined by seasons, then October is the sports equinox. It’s the only month where all the major sports in North America converge for your attention no matter the powers of deflection. Major League Baseball playoffs are held in October. It’s the middle of the football season for College Football and the National Football League. It’s where Major League Soccer concludes it’s regular season to begin it’s playoffs for the MLS cup. It’s also the start of Hockey and Basketball seasons for the National Hockey League and National Basketball Association. October concludes the Women’s National Basketball Association seson where the world famous Los Angeles Sparks won their third WNBA Championship last week (as of this post writing). You can see why October is a massive month for North American Sports and for yahoo, Fantasy Sports is one of their remaining profitable products.
October is the most active month for Fantasy Sports. It’s the middle of Fantasy Football season, so if your team is 0-7 at this time of writing, just dump your team to the waiver wire and prepare for next year. It’s also the start of Fantasy Hockey and Fantasy Basketball drafts. The Fantasy Sports Trade Association website estimates 57 million fantasy sports players play fantasy sports in the USA and Canada alone. It’s big business.
Basketball is my favorite fantasy sport out of all the major sports. I’ve played fantasy basketball for 15 years in many private leagues using yahoo as our provider. Some of my picks are either a hit or a miss. I could congratulate myself for drafting Anthony Parker because he was a “sleeper” pick in 2006. He filled up the stats with the Toronto Raptors after playing for Maccabi Tel Aviv BC’s Euroleague championship run. On the other hand, Dwight Howard was my first pick when he joined the Los Angeles Lakers posting mediocre stats for an all-star player in 2012.
I didn’t count injuries as a factor. Injuries are risks professional players take as the season progresses. Average Draft Position (ADP) yields owners’ projections of injuries as factors of a player they choose to draft.
What’s an Average Draft Position (ADP)?
Average Draft Position renders when you can expect a player to be picked across a fantasy sports landscape.
Let’s say you have the 3rd overall pick of your fantasy basketball draft. The first top four players in your pool are, in order: James Harden, Karl Anthony-Towns (KAT), Stephen Curry, and Russell Westbrook. The ADP measures the picks from an aggregate set of data based on previous fantasy basketball drafts at the start of the season from last year’s results and the current season’s projections. ADP accumulates the results of picked players from a number of leagues: Harden at 1st, KAT is either the second or third pick, Curry is the second or third or fourth pick, and the picks are accumulated by each fantasy basketball manager until the draft is over. I drew the table below to illustrate the example (fantasypros or yahoo or any rankings vary their ADPs, but the players are ranked as is).
| Pick | Rank | ADP | Player | Position |
|---|---|---|---|---|
| 1 | 1 | 1.0 | James Harden | G |
| 2 | 2 | 1.3 | Kevin Durant | G/F |
| 3 | 3 | 2.1 | Stephen Curry | G |
| 4 | 4 | 2.3 | Russell Westbrook | G |
| 5 | 5 | 2.9 | Karl Anthony-Towns | F/C |
ADP is the equivalent of a “buy or hold” rating if you’re going to buy shares of a company in the stock market. You use the average data to pick a player in a snake draft as your window of opportunity to draft that player you want. ADP isn’t useful for auction drafts since you have the opportunity to bid on every player on the bidding block. ADP is best used for Snake drafts.
How you use ADP depends on your drafting strategy. Points and percentages are data dependent for your team’s mission of strength. If you chose to specialize in Rebounds, Blocks, and field goal percentage, you draft Power Forwards and Centers. The same can be said for Assists, steals, and 3 point percentage for Guards and Small Forwards. Average Draft Position could help you pick players for specialized stat filling as the round robbin progresses in the draft. The closer you chose a player to the final round of the draft, the more likely you’ll use ADP to pick a “sleeper”. Sleepers are players who exceed expectations.
So something caught my eye last weekend when I saw Julyan Stone above Javale McGee in ADP rankings.
I want to break this down before I spend an entire week (I have free time in Denver) looking at fantasy sports bots.
Did Julyan Stone benefit from fantasy bots?
Most bots are designed to monitor activities from websites, messaging forums, purchases, bids, anything to accumulate data to trigger point to point ping responses. If i’m building a reddit bot, I’m going to program it for the following:
- Read a Post from Reddit
- Reply to Posts
- Automate the Bot for replying to posts
If I’m going to program an ADP bot for fantasy drafts, I’m going to use it for accumulating results of a draft:
- Scan Draft to meet team requirements per league.
- Gather final results after league drafts end.
- Activate algorithmn for averaging player picks with round picks.
Is there a method for that madness? Is somebody trolling us to believe Julyan Stone is the real deal? Is somebody creating bots to increase his value in fantasy basketball despite the fact he was sadly relesaed by the Indiana Pacers before the season began?
Why would bots help fantasy leagues?
Fantasy sports players vary from die-hard sports fans to casual fans who play with the data for fun. They want to have an edge over their opponent and it’s a competitive environment providing trash talk and boasting against their opponents. They use the waiver wire to pick their “flavor of the week” as an edge to win. Victory is the orgasmic satisfaction, especially in fantasy football.
Microsoft and Yahoo built fantasy football bots for next year’s fantasy football season. It’s an experiment of comparing players based on news, tweets, stats, and rankings. It’s the comprehensive method for fantasy owners’ central operation to get everything they need in making decisions within 5 seconds that used to take 5 minutes or 5 hours to research a player from different fantasy sports websites like fanduel, fantasypros, and the list goes on. I personally know people who play fantasy football to spend up to 5 hours on a quarterback if he’s going to give him 300+ yards and 3+ touchdowns in a weekly basis. It’s nuts.
That’s why bots would help players in fantasy leagues. You don’t want to waste 5 hours debating if Matt Ryan is a better quarterback than Phillip Rivers. Bots would build the research, you just spend a minimum of 5 seconds to make a decision from your smartphone.
Stop picking on poor Julyan
I saw his preseason games in Las Vegas and he’s a great player. He’s a professional basketball player who will be in a team this season. He stood out in the drafts this year for the wrong reason and that’s why he’s the case study for proposing better fantasy sports analytics.
Who in yahoo decided it was a good idea to weigh mock drafts equally with actual drafts for ADP ranking? I don’t think these errors occur by accident. A myriad of possibilities led me to draft Julyan Stone as the 11th pick in one of my leagues based on speculation. None of which made sense, but I gambled with it.
Now i’m drawn into it because this gave me more questions behind Yahoo’s rankings or even taking on Yahoo’s scale of operations. How many of those yahoo compromised accounts are bots using crawlers for a purpose? Could it be sabotage? Could it be a marketing gimmick by his sports agent?
Whatever the case, I think these kinds of glitches warrant better policies for improving fantasy sports leagues.
Javale McGee on the other hand, well at least he’s getting better.
Is Average Draft Position Bullshit?
Consider ADP’s algorithmn still includes players who are fighting for that final roster spot otherwise they’ll be released or sent to the development league, like Julyan Stone. ADP also includes players who are retired or on the verge of retiring due to injuries, like Chris Bosh of the Miami Heat.
I say this for two reasons. First, ADP data is accumulated from a percentage of players who make rational or irrational decisions. As new players enter the sport from a previous fantasy sport, the chances of them making irrational decisions is highly probable. Experienced fantasy football players brand those owners “The Taco”, an owner who has no idea how to play. Second, yahoo accumulates ADP by combining draft results from Actual and Mock drafts and there’s no telling if the rankings are highly speculative or educationally useful.
If yahoo did the latter, which I’m speculating, then that’s a problem.
No wonder why I drafted my basketball team poorly this year. ADP was penetrating my skull. Add that to a list of problems yahoo should fix before they fix their security breaches and permissions of security bidding for ad space in their advertisement algorithmns.
Furthermore, I don’t know if someone fiddled with Yahoo’s REST API with a bunch of crawlers to troll us (okay me) to promote up-and-comer NBA players. If that’s the case, then I have a lot of homework to see how it can be done for fantasy league sabotage. Fantasy Baseball is coming up and drafting a Single A ballplayer as your last pick in a league that doesn’t allow minor leaguers based on your commissioner’s rules depends on the lols to give.
Or maybe I just suck this year.
More to come,
Al