A small but apparently vocal number of you remember that I used to write weekly columns each fall about the national challenge fantasy games. By very slightly popular demand we have decided to revive the column for one and only one appearance this year.

Below is a quick refresher on various bits of strategy that are most directly applicable to two games run by CDM Sports: the rotisserie- and points-scoring versions of the Football Challenge.

DISCLAIMER: This column pertains only to one peculiar kind of fantasy football. When I suggest that Trevor Siemian’s game last night was exactly as useful as Cam Newton’s, the analysis hinges on the fact that each player carries an assigned salary in these games, and Newton is hugely more expensive than his far less famous counterpart. If you treat anything written here as if it applies to more traditional fantasy formats, you are probably not helping yourself.

The basics:

Know the game you choose to play.

In the rotisserie version of the Football Challenge you must build a carefully balanced roster; when two of eight scoring categories are passing yards and passing average (yards per pass attempt), and you own half as many active quarterbacks as running backs or wide receivers, it rarely pays to skimp on passers. Plenty of teams will save salary in Week 1 with the likes of Kirk Cousins, Jameis Winston and Derek Carr, but if you own Dak Prescott you may have him to yourself. That can be good or very, very bad.

In points scoring you need only aim for raw accumulation. A point is a point no matter which position it happens to come from, just like in traditional formats—and top quarterbacks here are often dramatically overpriced relative to players like Prescott and even Siemian. The latter was not great last night; his debut came to 13.9 points by CDM’s reckoning, and Newton notched 24.1 after pacing all players in 2015 with a weekly average of 26.3. So Siemian is something like 55 percent of a top points quarterback right now … which is actually not so bad when you consider that Newton costs $5,310 against a salary cap of $60,000, and Siemian costs only $1,110. You should almost certainly not own Siemian, but you can maybe skip Newton, too.

Likewise, in points there are many better ways to spend your money than by springing for a top kicker, but in roto kickers get one category all to themselves (a running back can catch passes for 47 yards and a touchdown, but no non-kicker will ever chip in a field goal) and make significant contributions to another. Wil Lutz makes a fine speculation in points. Given the Saints’ recent history of coughing out fewer kicking points than the quality of their offense might predict, I am trying hard to keep him out of my Week 1 lineup in roto.

The schedule matters …

Which should go without saying, but plenty of entrants will take too short a view of things when locking down their Week 1 rosters. I mentioned that teams will save salary with Winston? Perfectly defensible this week, as the Bucs head to Atlanta to take on a pass defense that Football Outsiders ranked as 22nd-best in the league in 2015. But the same Bucs have an early bye this season (Week 6), and their other opponents between now and then are Arizona (4th), Los Angeles (8th), Denver (1st) and Carolina (2nd). Oof.

… but the schedule does not matter that much.

Then again, 2016 is not 2015. Carolina built last year’s ranking in significant part on the brilliant work of Josh Norman, Los Angeles on the work of Janoris Jenkins, Arizona on the work of a fully-healthy Tyrann Mathieu. As many things change from one season to the next in the NFL, it will be a few weeks yet before we know what is what—and meanwhile Winston could in fact be en route to a big sophomore season.

It always amuses me to see entrants telling the CDM bulletin boards how their rosters are set for the first several weeks of the season: schedules perfectly accounted for, appealing matchups for everyone, just need to shuffle between active roster and taxi squad, no need to burn any early-season purchases. Right. I have been playing these games for twenty years now. The number of times all has gone according to plan in that span is zero.

Making good use of the taxi squad can mean lots of different things.

In those twenty years I have never gotten my bench exactly right in a challenge game. Some years I stack the taxi squad with stud players I cannot really afford to have active, figuring that as basement bargains emerge I can buy them one at a time but be adding pieces two at a time as I do—Prescott comes on, and he clears enough salary space to bring Todd Gurley or Antonio Brown with him. Some years the taxi squad has more of a stars-and-scrubs look, with Prescott already there alongside Gurley and Brown. The problem with the former approach is that you have to drop someone when you add someone else, and a bench full of studs makes for painful in-season cuts. The problem with the latter approach is that it is all but impossible to anticipate which scrubs will emerge as playable options. In roto, where the salary cap is $30,000, I can tell you with some confidence that the most useful piece that could emerge in 2016 would be a playable cheap quarterback. What I cannot begin to tell you is whether “playable” might eventually describe someone in the $1,400 salary range (Robert Griffin III, Blaine Gabbert, Jimmy Garoppolo) or someone who saves considerably more (Prescott costs $700, Siemian $620).

And this year especially there is another factor to consider, as several top talents (Tom Brady, LeVeon Bell, Josh Gordon) are suspended to start the season. There are good arguments to be made both that the dumbest thing you can do in these games is burn a purchase on a player you knew you wanted all along, and that the dumbest thing you can do in these games is roster a player who cannot possibly help you for a month.

Cheaters always prosper.

That has been the company line for as long as I have worked here, and it is never truer than in context of the challenge games. CDM allows penalty-free signups all the way up to a few minutes before the first kickoffs this Sunday—and by penalty-free, I mean that everyone gets a free crack at the numbers from last night’s “cheat” game. You can stock your active rosters with as many Broncos and Panthers as you wish.

Unfortunately, there is almost no edge to be gained from the cheat game this year. I had hoped that Virgil Green would catch more passes, but Carolina had one of the stingiest defenses in the league against opposing tight ends in 2015, and but for a wide-open end zone look that was broken up, Green had a predictably quiet opener. All factors including salary considered, I think the whole universe of playable cheat gamers consists of C.J. Anderson (must start in both games) and Kelvin Benjamin (should start in both games).

And that is all I can tell you right now. I never lock down my own rosters until a few minutes before the deadline on Sunday; year after year I seem to wake up that last morning with some brand new idea having come to me in a dream, and I spend the morning trying frantically to decide if dream me is any smarter than the conscious version. I wish I could defend one or the other of my selves by telling you a clear pattern has emerged, but after twenty years the results are still mixed.

Good luck in 2016. As I used to write every fall, my fondest hope is that one of my readers will finish in second place overall in each of the challenge games. First place, of course, is spoken for.