Some might look at the 95-67 Rangers, who only scored eight more runs than they allowed, or the 84-78 Yankees, who were outscored by 22 runs, and call these teams lucky. That’s a fair opinion and one that’s backed up by the Pythagorean Theorem of Baseball. But it’s not definitive, and every year there’s going to be teams that over and underperform.

As the old saying goes, “It’s better to be lucky than good.” There may be some truth to that, particularly when posing the question, “Which teams offer their bullpens the most save opportunities?”

Conventional wisdom states you want to target the bullpens of the teams that win the most games since having a lead is a requirement for recording a save. Others claim that teams with low run differentials offer the most save opportunities since the games they do win are usually going to be close enough for the chance of a save.

There’s plenty of opinions on the topic, but what stats have the closest correlation with save opportunities? Here are some linear comparisons of team statistics to save opportunities:

As it turns out, win percentage does seem to be the closest indicator of save opportunities. However, an R-squared value of 0.16 means that just 16% of the variation in save opportunities can be explained by a team’s record. We can gather from this that winning teams have a slightly better chance of offering their closer saves, but it’s nothing close to a definitive answer. Runs allowed was surprisingly the second best predictor, with an inverse correlation to save opportunities, although it’s close to the territory of insignificance.

Run differential had a direct correlation with save opportunities, but there’s not a whole lot here either. I also used the absolute value of run differentials to test the hypothesis that teams with run differentials close to 0 found more save opportunities.

This did not seem to be the case, as it matched up with the plot of save opportunities almost 50% less than run differentials alone. And finally, runs scored saw almost no predictive value. How about combining a few of those stats together, though?

That’s a little better. The multiple regression of win percentage and run differential saw the closest line to the plot of save opportunities of any combination of two stats from the table above. It’s worth noting that run differential has an inverse correlation this time.

If it makes sense that teams who win a lot of games by small margins are more likely to see save opportunities, the stats back it up. This takes us back to the ‘lucky’ 2016 Rangers and Yankees, who both outperformed their expected win rates. We’re still only explaining about 24% of the variance, but it’s a better bet than win percentage alone, and certainly more so than any of the other statistics in the first table.

The takeaway here is that no one stat is a great predictor of save opportunities, but this combination is a place to start. Here’s the entire chart of expected save opportunities from last season, using win percentage and run differential:

There are a few caveats to interpreting this chart. Firstly, save opportunities aren’t limited to the ninth inning – a blown three-run or less lead in the eighth inning counts as a blown save. Closers don’t inherit 100% of save opportunities in the ninth anyway, so a 70 save opportunity team won’t realistically have a 60+ save closer.

Secondly, this chart is descriptive of statistics from last season, but not necessarily predictive of future statistics. It may be more worthwhile to reexamine using data mid-season or modify based on projections.

Run differentials are the best predictors of win rate over the long term, but there may be something more going on with teams that vastly outperform theirs in a given year. That being said, let’s take a look at some interesting RPs going relatively late in drafts who are inheriting good team situations:

Kelvin Herrera (117.0 ADP, 10th in team xSVO), Francisco Rodriguez (137.3, eighth), Sam Dyson (166.0, first), Shawn Kelley (208.3, ninth) and Jeanmar Gomez (335.0, fifth) all have something in common. They could all be closing for teams that ranked in the top 10 of expected save opportunities last season. Kelley and Gomez are likely falling in drafts since Kelley faces contention for his closer role with Blake Treinen and Phillies Manager Pete Mackanin just announced that Jeanmar Gomez “deserves to be called our closer,” so his draft stock will be on the rise.

Kelley (12.41 K/9, 1.71 BB/9) put up elite rates last season, so he could be a value pickup if the Nationals don’t add another late-inning relief pitcher to their bullpen. Gomez is coming off an unimpressive 4.85 ERA and 3.96 FIP, so Hector Neris (11.43 K/9, 3.36 BB/9) should be monitored.

The other three look to have relatively safe job security, so their draft positions are the result of other questions. Herrera is a favorite draft target of mine in his first full season as Kansas City’s closer. With a strikeout rate of 10.75 K/9, a walk rate of 1.50  BB/9, and a 2.47 FIP, he appears poised for a big season.

Pitching for the Royals promises an elite defense, a park that suppresses HR, and plentiful save opportunities. His 15.2% swinging strike rate ranked 12th among RPs, which legitimizes the jump in strikeout rate from 2015 to 2016.

Managers might be deterred by Francisco Rodriguez’s age (35) or mediocre rates (8.02 K/9, 3.24 BB/9), but he still saved 44 games last year. I’ll be joining those staying away due to the decline in strikeout rate and spike in walk rate in an older player.

The warning signs are present, but for the bold, Rodriguez could be a strong source of saves if he can hold onto the closer role all year. Detroit has other viable options in Bruce Rondon, Justin Wilson and others if Rodriguez stumbles.

Sam Dyson took over as the Texas closer in the middle of May and went 38/42 in converting save opportunities with a 2.43 ERA. But similar to Rodriguez, he’s being drafted as the 22nd RP. Also like Rodriguez, Dyson’s 7.04 K/9 and 2.94 BB/9 place him well below the top tier RPs, but ERAs of 2.14, 2.63, and 2.43 over the last three seasons while maintaining similar peripherals indicate an ability to suppress hard contact.

His 65.2% GB rate ranked third among RPs behind only Zach Britton and Blake Treinen. I’m still wary due to the below-average strikeout and walk rates, but the Rangers represent an ideal situation for saves and Dyson may be flying too far under the radar right now.

The teams of these pitchers might not rank in the top 10 of expected save opportunities again next year since so much changes year-to-year. But it’s a place to start looking, and there is a real correlation here.

So keep an eye out for next year’s ‘lucky’ teams. They may just help you pick up some saves.

Subscribe: iTunes | Stitcher | SoundCloud | TuneIn | Google Play

Alex Isherwood is a correspondent at FantasyPros. For more from Alex, check out his archive.