Identifying Risk via ATC Projections – Top 100 Players

Ariel Cohen uses his ATC fantasy baseball projections, the most accurate for two straight years, to discuss risk among the top 100 players for 2022 MLB drafts.

In my previous article, “Identifying Risk via the ATC Projections – Round 1,” I provided a primer on the ATC Projections volatility metrics.

To further your understanding, and for a visual explanation of how these volatility metrics work, please check out the seven-minute video entitled, “Using ATC Projections to Minimize Fantasy Baseball Draft Risk.” This was done in collaboration with fellow RotoBaller, Eric Samulski.

The first article of this series previewed players currently drafted inside the first round of NFBC drafts. Today, I will take a closer look at players being selected in the top-100. Note this article was originally published on March 1st.

• Fantasy baseball trade analyzer
• BvP matchups data (Batter vs. Pitcher)
• PvB matchups data (Pitcher vs. Batter)
• Who should I start? Fantasy baseball comparisons
• Daily MLB starting lineups
• Fantasy baseball closer depth charts
• Fantasy Baseball live scoreboard
• Fantasy baseball injury reports

ADP Data

To set the stage once again for this exercise, below is a visualization of recent NFBC ADP data for Draft Champion leagues:

Click image to enlarge

Player positions in the above grid are colored by the following key:

The grid will enable us to locate and analyze players by round when we talk about risk in the coming sections.

InterSD Player Notes – Top 100 Players

Below is the same NFBC ADP grid, but rather than the positional display - players are colored (and shaded) by their ATC Inter-projection standard deviation (InterSD). Players colored in green have low InterSD values, while players colored in red have high ones.

Click image to enlarge

To remind the reader - InterSD is a proxy for a player’s parameter risk. It quantifies how differently ATC’s underlying projections are from one another. My research has shown that players with high levels of parameter risk tend to underperform projections on average and vice versa.

We previously discussed players in the first round of drafts, so let’s continue now with the rest of the top-100 players.

In the second through seventh rounds, the players with the lowest parameter risk are Matt Olson (1.73), Nolan Arenado (1.74), and Atlanta’s relief pitcher, Will Smith (1.77).

Olson’s first year with substantial playing time was his age-23 season in 2017. Despite only having 216 plate appearances that year, Matt swatted an enormous 24 home runs. That tremendous pace was certainly unsustainable for a full season, as it was propelled by a 41% HR/FB% ratio. The power figures regressed over the course of his career but remained far higher than the league average. Since 2019, Olson has hit no less than 36 HRs per season (or of that pace in the shortened 2020) – and all of the supporting analytics show that he is fully legitimate. In 2021, Olson improved his game further by lowering his strikeout rate to 16.8% and putting up a career-high 5.0 WAR.

For 2022 - no matter what projection system you look at, Matt Olson is slated to club homers in the high 30s. For all systems, his projected auction value in a 15-team 5x5 mixed roto league lies between $20-$24. If anything, as seen in his ATC InterSK value (-0.50), he could even beat these projections. Simply look to 2021, where he amassed $28 of roto value - a season that he could replicate.

Nolan Arenado’s transition to St. Louis went largely as expected. He no longer is the $30+ fantasy asset from a few years ago, but he still is highly dependable as a fantasy third baseman. His 85% contact rate in 2021 was largely in line with his career norms and sets the stage for a high-value floor. Yes, his HR/FB% rate is no longer the same figure while back in Colorado, but that is already accounted for in his 2022 projections. With projections in high agreement this year, and with third base presenting as a relatively weak position in drafts, consider rostering Arenado on your fantasy teams.

As for Will Smith (ATL, RP) - it is important to note that projections typically agree more of the time for hitters than for pitchers. The fact that Smith has the third-lowest InterSD of rounds 2-7 speaks volumes about his low parameter risk.

The main reasons that he shows up with such a low InterSD are:

• Almost all underlying systems project him to close out an exceptionally large percentage of Atlanta’s games. The saves account for a large part of Smith’s fantasy value.
• Innings and strikeout rates are very similarly projected across systems.

Smith is still going relatively high in NFBC drafts – in the late fifth round. But with his indicated parameter risk metric, he might be a better buy than spending a second-round selection on a (pricy) Liam Hendricks.

Note: This article was written before Kenley Jansen signed with the Braves, bumping Will Smith from the closer's role. 

On the other side of the coin, players with high InterSD values typically fall into the following three categories:

• Players with injury risk
• Players with little major league experience
• Players with an uncertain role

Of note, the players with a large amount of ‘injury risk’ heading into 2022 are mostly pitchers. Justin Verlander (9.1), Jacob deGrom (7.3), Shane Bieber (7.3), and Chris Sale (6.9) lead the field in pitcher InterSD. None of these names should surprise anyone.

Bobby Witt, Jr. (7.7), one of the top prospects in Major League Baseball, falls into the category of ‘little major league experience.’ His name here too should not surprise anyone. The major contributor to his parameter risk (and others in this category) is the projected playing time – which is often hard to predict for prospects. Witt could play a full season, half-season, or nothing at all.

Note: In the most recent update of ATC Projections, underlying playing time estimates (and other categories) have gotten closer to each other for Bobby Witt, Jr. His corresponding InterSD remains high, although it has since dropped to 4.8.

Kenley Jansen (6.5) falls into the third high InterSD category – of uncertain role. Currently, Kenley Jansen is an unsigned free agent. He may wind up being the closer on the next team that he plays for … or he may be just a high leverage reliever. Projected saves totals for him range from 45 saves all the way down to … zero. He is a prime example of an uncertain true talent for the coming year.

Note: This article was written before Kenley Jansen signed with the Braves, bumping Will Smith from the closer's role. 

InterSK Player Notes – Top 100 Players

Below is the same NFBC ADP grid, but this time, players are colored (and shaded) by their ATC Inter-projection skewness (InterSK).

Click image to enlarge

To remind the reader - The ATC InterSK quantifies how the underlying projections differ … not just if they differ. Positive-skew occurs when the median of the distribution is lower than its average. It is when more of the underlying projections are found lower than the ATC average. Negative-skew is the reverse. It is when the median of the distribution is to the right of the average – when more projections appear higher than the indicated ATC average.

The negatively skewed player distributions (above) are colored green, while the positively skewed ones are colored red. Negative-skew is preferred to positive-skew (green-colored players are preferred to red ones). All things being equal, you should want to draft a player where the wisdom of the crowds is better than the ATC average.

The players with the most negatively-skewed distributions (the most negative InterSK) in rounds two through seven are Cody Bellinger (-2.5), Chris Sale (-2.3), Aaron Judge (-2.2), and Shane Bieber (-1.9).

As detailed above, Sale and Bieber both have very large InterSD values. Cody Bellinger has a large InterSD (4.2) as well. Paired with the negative InterSK, it means that projections differ greatly, but the wisdom of the crowd skews above the ATC average. That should somewhat temper the extreme parameter risk of the three in question.

Aaron Judge, on the other hand, has a small InterSD (2.8). Coupled with a very negative InterSK – Judge’s parameter risk component is even smaller. Projections tightly agree, and if any – outlier projections are low.

For example, Judge’s lowest HR projections come in at 30. The ATC average lies at 38, and the median HR projection is even greater at 39. The nine HR spread between the observed 0^th percentile and the 50^th percentile is particularly large, given the fact that the bulk lies between 38-40.

Yes, another injury can certainly occur with Judge … and yes, there is always the possibility of prolonged underperformance. But with the low underlying ATC volatility parameter risk metrics, we can assert that Judge’s true underlying power is that of a high 30’s HR output. Go the extra buck (or half-round) for the Yankee slugger in 2022 drafts.

Now onto the players with very positive InterSK values in the top-100.

Kris Bryant (+1.6), Yu Darvish (+1.5), Lucas Giolito (+1.3), and Luis Castillo (+1.3) exhibit the most positive skew in their underlying projected true talent distributions.

For Darvish and Castillo, these highly positive InterSK values caution us that full bounce-back seasons may not be in the cards. For Giolito, one should be more cautious of his two-year breakout. Though the three pitchers’ InterSD values are fairly neutral – the indicated large positive skew points to some added parameter risk.

For Kris Bryant, the InterSK flag is far less worrisome, with his strong 2.7 InterSD value. Consider his projected home run totals.

Bryant’s low projection is 21 HR. The vast majority of projections are between 23 and 24 HR. Yes, Bryant does have a 30-HR outlier projection, which causes the positive skew. However, that flag is heavily outweighed by particularly strong convergent projections on the whole. No matter what the exact shape of a tightly knit distribution is, it is nonetheless tightly knit.

Bryant is still about a round overvalued, according to ATC, but his risk metrics do not add any additional flags to his prospective value.

IntraSD Player Notes – Top 100 Players

Finally, below is the NFBC ADP grid once again. This time - players are highlighted by their ATC Intra-projection standard deviation (IntraSD). Low IntraSD figures are shaded in blue, while high IntraSD figures are shaded in orange.

Click image to enlarge

IntraSD is a measure of how spread out a player’s value is over all of the categories. The intra-projection volatility deals with the player’s categorical or profile risk. The smaller the categorical risk of the player, the less effect that any single component of the player’s profile will have on his total rotisserie value.

For pitchers, the high IntraSD players are typically closers. Their value is concentrated within the saves statistic. For that reason, Josh Hader (2.0), Liam Hendriks (1.8), and Raisel Iglesias (1.7) all have high IntraSD values. Of course, closers cannot be ignored in drafts – and these are the best of the bunch. The large IntraSD values simply highlight the extra risk inherent in their profile.

As for hitters, we typically see the ‘speedster’ and ‘home run hitter’ type players generate large IntraSD values. Adalberto Mondesi (2.9), Starling Marte (2.2), and Whit Merrifield (1.8) are some of the highest IntraSD speedster hitters. Pete Alonso (1.4) and Yordan Alvarez (1.3) are some of the higher IntraSD homerun hitters. As opposed to closers who are the only sources of the saves category, for hitters – you do have other choices to fill up your aggregate SBs and HRs. The IntraSD simply highlights the added profile risk. You might want to roster these smaller dimension players at a slight discount.

Now onto the low IntraSD players. Of the hitters selected in the second through seventh rounds, J.T. Realmuto (0.29), Jorge Polanco (0.32), Kris Bryant (0.33), Christian Yelich (0.43), Ozzie Albies (0.44), and Byron Buxton (0.47) have the lowest IntraSD values.

A few notes about each of the above players:

• J.T. Realmuto’s categorical spread is close to uniform. Realmuto is projected for a slight positive Z-Score in each of the five rotisserie scoring categories. Particularly because he is catcher-eligible, he is highly unique within the player pool.

• Not only is Jorge Polanco’s profile well balanced, but he is going at a discount, according to the ATC Projections. With a late sixth-round ADP, the Twins’ middle infielder may be a helpful piece to your roster’s construction at an accretive price. To boot, his InterSD is only 2.0!
• We have already discussed Kris Bryant. With a very low 0.33 IntraSD, add one more reason to roster the former rookie of the year.
• It is hard to say whether Christian Yelich will bounce back to his former self. It is possible that he underperformed due to injury over the past two seasons. It is also possible that he still remains injured or in decline. But at a late seventh-round price tag, ATC volatility metrics say that he is safer than most.
• With a 2.4 InterSD and a 0.44 IntraSD, Ozzie Albies is quite stable. However, compared to the NFBC ADP – he is overpriced. The market clearly already knows about his rock-solid profile.
• Byron Buxton is a tough player to properly price. With a low profile risk as highlighted by IntraSD, he is a tremendous building block when constructing a roster. However, projections do not agree on his values (5.2 InterSD). The disagreement stems mostly from playing time. There are also a few rates that are highly dispersed. One, his projected slugging percentages are all over the place - ranging all the way from .450 to .650. Second, his stolen base projections are also divergent. His fourth-round price tag might not be low enough to earn a profit. However, if he drops any further, take a chance on his excellent profile.

Download Our Free News & Alerts Mobile App

Like what you see? Download our updated fantasy baseball app for iPhone and Android with 24x7 player news, injury alerts, sleepers, prospects & more. All free!

More Fantasy Baseball Advice