Points League Strategy and Tips

So, you’re in a points league.  Nice!

You don’t have to trouble yourself with individual ‘categories’ that everyone else is talking about, or those pesky things called ‘ratios’.  You have one statistic to worry about (the points you score) and want to find the best formula for maximizing it.

Now, many people might think points leagues are for the casual fan who craves a simpler format, the pedestrian fan who is bored until football season returns.  That’s not true.  Even though you have just that one statistic, a lot goes into figuring out how to draft the best points league team possible – as you’ll see.

Before proceeding, know this is a deep dive based on a system developed over a decade ago.  It still works pretty well.  But before we get started, you need to…

 

Know your scoring system

 

If you’ve read some of our previous dissertations on points leagues, their best feature is that you’re not beholden to a limited set of stats.  Points are typically added for walks and extra base hits, but you can include a half-point or so for a guy who takes an hit by pitch for the team (he just took a 97 mph fastball to the back!  He deserves that half-point!) or has an outfield assist, whatever.  Same for pitchers – throw in some points for successful pickoffs or a penalty for wild pitches.  You can do what you want as long as there’s balance to it.

And no matter how extensive your scoring system is, you’d better know it through and through.  If it’s truly balanced, there shouldn’t be an easy way to exploit it – hitting isn’t overly weighted against the pitching, power doesn’t overwhelm speed, etc – sometimes it takes a season or two to find the sweet spot.  That said, there’s always going to be a cheat code if you look hard enough – after all, that’s what this article is for.

Next, know the exact number of roster spots you have – three outfielders or five?  Corner and middle infield slots?  How many bench slots?  All this is important, you’ll see why in a minute.

Homework time.  First things first, you’ll want to…

 

Establish your baselines

 

Once I’ve familiarized myself with the scoring system, the next thing I do when prepping for a draft is identify the average points that are typically scored by each roster spot.  Let’s say you have a super-basic breakdown of a basic MLB team – one spot for each basic position, five or six pitcher slots, and a handful of bench slots (which should only be occupied by starting pitchers, by the way).

At the start, you want to establish a baseline for the points needed to be at least an above-average team.  Here’s how you do it:

First, go to last year’s season in your league.  You can also do this over multiple historic seasons to get a better sample size, but let’s take this one at a time.

Next, go through each position and sort by points scored.  You will want to copy/paste the totals of the top finishers into a spreadsheet.  The number of players you pull will depend on how deep your league is, but in my example let’s assume it has twelve teams and just those basic positions, so do the first twelve for each (if you have three outfield slots, do the top 36; if five pitcher slots, do the top 60).

Try to ignore the player names for now, though you want to keep them handy for later.  Some things might already be jumping out at you, like that Matt Vierling was a top-12 third baseman in ESPN standard points leagues, and he scored the same points as Julio Rodríguez in fewer at-bats (will that happen again?  Probably not, but put it in your pocket).

Once you have your point totals, you want to do some basic spreadsheet ninjitsu to get the following:

1. The top scorer at each position
2. The average point total of the top <however many teams are in your league>
3. The bottom scorer of the top <however many teams are in your league>

So for that 12-teamer sample with a basic roster, I want to take the top twelve catchers, first basemen, second basemen, etc.  Here’s what my hitting spreadsheet might look like (again using ESPN standard scoring):

Keep going until you’ve done them all.  For outfield and pitching, you’ll want to do a row for each set of twelve (or however many teams you have) so you have something like this:

Once you’ve done this for every position, what does this get you?  We’re far from done here, but you now have a barometer for each roster slot and how many points you can expect from them for an ‘average’ team.  You know what you should get from your ‘free’ players at the end of the draft by position and what to expect if you reach for something in the early going.  This will not only help you draft effectively but also help you evaluate your team as the season goes on.

Of course, what you’ll notice immediately is…

 

Not all positions are the same

 

Blatantly obvious, I know.  But in this ESPN league’s format, the average top-12 catcher will score about 160 points less than the average top-12 outfielder (or around a point for every game of the season).  That’s a huge divide.  That said, a catcher you expect to score 350 points in a season JUST MIGHT be more valuable than an outfielder who scores 500.  Let’s not get ahead of ourselves, just throwing it out there.  Another thing you’ll see is…

 

Deeper positions have diminishing returns

 

Let’s take a look at the starting pitchers:

The further down you go, the more ‘pooling’ there is – the variances are smaller from roster slot to roster slot.  The same is true with outfielders – only 44 points separate your average OF2 and OF3.

By now, other things might jump out at you, but put a pin in that for now.  It’s time to evaluate some players.  When doing this, you’ll first need to…

 

Know what you can predict

 

Your league might have a lot of scoring categories (again, the more the better).  But however many categories you have, all those stats amount to a single total point value for each player over the season.  Here, you want to be better than your league at predicting these totals.

Rather than using your platform’s projected totals (which usually just takes last year’s stats and reverts them to the mean in some degree) or rankings (irrelevant up to this point), you want to come up with your own.  Now, depending on how many players you want to evaluate, this might take a little more work.  But this is where the simplicity of a points league is nice to have – your projections are a recipe that needs but two ingredients.

For hitters, these are:

1. Average points per plate appearance (we’ll call this PPA), and
2. Expected plate appearances (call this EPA)

Your projected point total will be the product of the two.

You could also use ‘points per game * expected games played’, but it won’t account for a player’s potential movement up the lineup or if they were traded to a lesser team with less lineup turnover.

So of these two ingredients – or variables, if you want to be scientific – which can you predict?

The first one, PPA, depends on a player’s season-to-season performance.  That performance can vary: they could have a career year or fall off a cliff.  Maybe their back hurts, and they’re not telling anyone.  Maybe they started a new diet and are in the proverbial ‘best shape of their life’.  We can endlessly try to evaluate this (and we still will, to some degree), but all we have to go on is their track record and your own best guess.

The second ingredient (EPA) is based on three known factors: their position in the lineup, how often they get a day off, and the quality of the hitters around them.  Now, we don’t know if a player will get injured, sent down, or traded mid-season, but we’ll sidestep the proverbial ‘Random Acts of God’ for this exercise (though your own best guess can still play a factor).  Again, focus on what you can predict and not what you can’t.

Okay, we’re making progress.  Up next is…

 

Your point projection formula

 

As with the positional baselines, you want to begin by looking to the past.  Namely, what did they do last year?  (Or, if you have more time on your hands, the last few years?)

Fortunately, you’ve already copied down your player stats from your league’s previous season – or at least the player and their points total.  If your league stats include the number of Plate Appearances last season, great (at bats can be a substitute but it will discount your high-walk guys a bit).  If you don’t, a site like FanGraphs is a good resource.

So if you have your player’s 2024 point total and the number of plate appearances, all you need to do is divide the two and get last season’s PPA.  For a player like Juan Soto, this was:

Now, you’ll want to turn this formula around when predicting next season.  Let’s start with this:

Right here is where your gut wades into the equation.  And that’s okay, it’s no fun if the math just gives you all the answers.  But as I stated earlier, the PPA can vary season to season, impacted by any number of unknown factors.  So here’s where you make your best guess and decide if this year the PPA will be better or worse and by how much.

Still, we know a little bit: in 2024, Soto had a career year with 45 home runs that might not be repeatable.  He’s also a Met now, moving to a new park that won’t be quite as favorable.  He batted second all last year and will likely do the same, now hitting after Lindor (a vast improvement over Volpe or whatever the Yankees were leading off with) and before Pete Alonso (a step down from Aaron Judge)…and so on. You get the idea.

This is the ‘Best Guess Factor’, or BGF. Will the PPA go up or down, and by how much?  Here you can only make your best guess.  I don’t want to quantify it for you. Maybe you think Soto will be even better this year now that he’s got his long-term deal and is at peace, and that’s your choice.  But you should start with the historical numbers – what you do from there is up to you (Free Will and all that).

Exercising my own free will as an example, I’ll project that his PPA goes down by a few percent.  So I’m going to put a 0.97 into my BGF field:

Again, that part’s arbitrary.  And a 3% dip is a conservative assumption; go as high or low as you want – but if you’re going beyond the range of .9 to 1.1 you’d better have a good reason.  I save that for rookies who might break out or guys coming off serious injuries.

Next, we need that second ingredient: Expected Plate Appearances (EPA).  And this is the more predictive variable, right?  Soto had 713 plate appearances last year but the leadoff hitter for the Mets had 689 plate appearances in 159 games.  Soto’s addition to the lineup will probably juice that a bit as he gets on base as much as anyone else, but you should assume each subsequent spot in a lineup gets about 15-18 fewer appearances throughout a season.  So let’s say Lindor hits 700 this year; that means Soto behind him gets to 685. (you can also go to ZiPs or Steamer for their projections if you want to skip this step)

Now, you have all the factors to make your prediction.  Just multiply your three columns together for Projected Points:

This might not be exactly far from ESPN’s projection of 548, but small differences can matter, and there will be players with wider discrepancies (as you’ll see).

Now, you might be wondering…

 

What to do with pitchers

 

Well, pretty much the same thing.  Let’s take a look at Seth Lugo, a player highlighted in last week’s Points League Risers and Fallers.  Last year, Lugo had a career-best campaign with over 200 innings pitched, 181 strikeouts, and a 16-9 record.  These weren’t fluky numbers, either – he backed them up with a 3.00 ERA and 1.09 WHIP.

Let’s calculate his PPI from last year:

Now, we’ll turn that around to this year’s predictions.  Lugo is unlikely to repeat his career-best totals, but if you look at the year before, he scored 289 points in 146 innings for a PPI of 1.975 – and he spent some time in the bullpen that season.  Still, let’s say our gut factor comes in at .95 compared to last year’s stats.  We’ll also peg him down a few innings (Steamer has him at 190.1).

Well, that’s pretty freakin’ far from ESPN’s projection of 337.  Interesting.  Even if you applied Lugo’s 2023 PPI and gave him 180 innings, he’d come out at 356 points.

Cases like this should jump out at you.  This is what folks in finance call ‘arbitrage’ – it means free money.  But remember, you need to do these projections for every player in the draft pool before you know exactly who stands out at each position (it may take a couple of hours, but it’s worth it).

We’re almost there, but by now, you’re probably thinking…

 

Okay I just spent half my day on a spreadsheet.  What now?

 

You’ve got your projected point totals for all the players in your draft (or as many as you had time for).  Now it’s time for results.

Pop in your league platform’s default positional rankings or ADP by position for each player (or in a salary cap league, their projected auction value).  You should have your 12 first basemen, 36 outfielders, 60 starting pitchers, and so on (again, this depends on the depth of your league).  It should look something like this – for simplicity’s sake, we’ll start using fake numbers:

Next, compare those point totals to the baselines you established earlier.  To do this, 1) take the bottom-scorer point total for that position last year – this is your ‘replacement level’ number for a player that’s basically free at the end of your draft.  Then, 2) compare it to the average for the same position.

Here’s what I mean – let’s take second base.  The lowest scorer among the top 12 had 298 points, and the average among the top 12 was 350 points (a difference of 52).  That’s the range among the middle-to-bottom half of the position; let’s assume the range holds steady at the top half too (it probably won’t be but that’s kind of the point).  So now, 3) double it to get the full spread (52*2=104).  Now you have your baseline for the expected points you might get from the draft slot at the top of the position vs the bottom, and if you had a baseline for every slot in between, divide 104 by 12.  Now you’re assuming that each subsequently ranked second baseman will score 8.67 points less than the player drafted before him at the same position, relatively speaking.

Now, your positional rankings would include these baseline values.

It’s all coming together now, but there’s one last acronym to throw at you…

PAR = Points Above Replacement

 

Your points per position are determined – it’s an even spread across the 12 or so (however deep your league is) players at each position, including multiple groups for outfielders and pitchers.  You want this ‘even spread’ so you can find the players who defy the expected downward trend of rankings, even if there’s more of a downward-sloping bell curve from elite players to merely good ones.  And, of course, drafts won’t smoothly rotate from one position to the next – you may have a run of outfielders at the top or three to four shortstops in the top 10.  But that’s all well and good because you are about to know the relative value of every single player in your draft and can adjust accordingly.

Now you’ve got your projections, your expected draft slots, and your array of ‘Points per position’ based on the draft slot.  Again, it’s not an exact science, but we’re trying to get as close as we can.  From there, just calculate the difference between your player projection and the points per position to get what’s called ‘Points Above Replacement’ (PAR).

What you are trying to solve for here is your player’s ACTUAL expected value versus the value of the draft slot they’re being taken in.  Knowing the difference between these two numbers gets you that arbitrage.

We’re rounding third here and headed home.  But, you’re asking…

 

Is there anything else left to do?

 

Well, yes.  But here’s the easy part.  Take the players that graded out the best in PAR for each position and start planning your team.  Spread things out – pick one from the top three, somewhere in the middle, and at the bottom for each roster spot – a favorite choice for each major tier.  It should look something like this:

Even better, go back to your platform’s ADP and pick the 1-2 players within each draft round with the highest PAR over the other players in their position.  Now you have targets round by round that should statistically outperform the others.  If you trust your numbers enough, you will have no problem reaching for a player a couple of rounds early – and that’s fine (in this ESPN league, for example, I’d happily have Lugo as a top-25 starter).

Once your draft room is open, simply populate your queue with all the high-PAR players you selected, stay disciplined with your strategy, and snipe at will.

You might also wonder…

 

If you’re in an auction

 

This system was initially developed for auctions because you can use it to determine the actual value of every player in your draft.  Take your bottom scorer at each position and assume that player will go for a dollar.  Then, in addition to a ‘points spread’ between the bottom and the top of each position, there’s a ‘dollar spread’ as well.  So going back to second base in the example above, if the top-ranked second baseman is expected to go for $37, you have a spread of $36 from top to bottom at each position (or $3 for every 8.7 points in your spread).  You can take your projected total, use that same price-per-point calc at each position, and come up with each player’s actual projected value.

Compare each player’s pre-draft auction value on your platform to your calculated projection.  Now your PAR becomes ‘Price Above Replacement’ but works just as well.

All the guys with the high-dollar PARs go into your queue.  One thing to watch out for: your league-mates might bid up many of these same players past the platform’s value, so know your spending limits.  You know what every player is worth to you, and that’s a tremendous advantage – so if a guy is sitting there below your price target, but you hadn’t planned on taking them, there’s still a value there.

And with that…

 

You’re done!

 

This system might take some time initially and setting up formulas in spreadsheets isn’t the most fun you could have on a Saturday.  But you will go into your league draft having personally assigned a target value to every single player and identified the ones that statistically stand out among the rest.  All in a few hours of study.

If you like or hate the system, let me know in the comments!  And if you think you have a better one, sign up to join Pitcher List’s Community Points League, where I’ll be defending my title.  Best of luck!