This is the first time out in what I hope will be a monthly look at what would happen if I tried to make a little money on these picks. The odds on every golfer are easily found at various places on the Internet (click on "LINKS” above), and the event-by-event picks are right here to see.

I’m looking at this from three perspectives:

• First, what would happen if a bet was placed on the first pick from each event’s Hawley rankings.

• Second, what would happen if a bet was placed on each of the first four picks from each event’s Hawley rankings.

• Finally, what would happen if a bet was placed on each of the first four "value” picks from each event’s Hawley rankings. A "value” pick is one where there is a favorable difference in the odds found on-line and the "Observed Odds” from the rankings (i.e. the on-line odds are some number X-1, and the Observed Odds are Y-1 where Y is less than X, or favorable to the bettor).

Over a lot of events (more on the methodology below), here’s what I see:

• Betting on the first pick, the bettor can count on hitting a few winners per year, and this method is doing great in 2017 (125% profit). However, most wins are at low odds and this approach shows a loss of 13% in the large sample shown below.

• Betting on each of the first four picks, the bettor can get close to break-even. The sample below shows lots of winners and a profit of 84% if you consider only 2017. Over the longer haul, it produces a loss of 3%.

• Betting on the "favorable” picks, finally, appears to be a very tough way to try to make a buck. In 2017 this method produces a 1% loss. Over 128 events, it shows a 41% loss. The risk in this method may be even worse than it appears because almost a third of the total winnings in 128 events came on one bet, Pablo Larrazabal at 108-1 in the 2014 Abu Dhabi event. Larrazabal won by one stroke over Rory McIlroy and Phil Mickelson.

From a methodology standpoint, the on-line odds are determined by surveying two to four of the on-line sites and averaging the odds on each player. For instance, in this year’s Honda Classic, Adam Scott was 11-1 at Bovada.com, 11-1 at Sports.ag, 12-1 at Sportsbook.com, and 9-1 at SportsInteraction.com, for average odds of 10.75-1. The Hawley ranking odds were 18-1, thus Y is greater than X and there is no statistical advance in selecting Scott.

These are hypothetical bets and what we’re looking at is the rate of return, i.e. the size of the bet doesn’t affect the rate of return.

I have been accumulating this data for a few years (not every event every year, but a lot of them). What’s shown here is results based on the last 128 U.S. or Euro events for which I have the data.

Notes: From a practical perspective, I assume $10 win bets at the given odds. In terms of summarizing the data, the number of 128 events is chosen because it’s the largest power of 2 (2 to the 7^th) that is less than the number of accumulated events.

Here’s the data for the last 128 events that I tracked.