Study The Sharks: Reviewing Past Slates with The SaberSystem

Welcome to the fifth and final episode of the Saber System. My name is Jordan Chand. I'm the head coach here at SabreSim. And in this series so far, we started by laying out the five fundamental principles you need to be successful in DFS. In other words, the Saber System, they are in order. First, understand the odds and the variance of DFS.

Second, optimize your DFS portfolio. Third, keep your lineup building process simple. Fourth, stay on top of the news, and finally, fifth, study the Sharks. In the subsequent videos in the series, we went through and entered our contest for a day of baseball. We built our lineups for the main slate and for one of the showdowns.

We did our late swaps, and now in this episode, we're going to be focusing on reviewing our lineups to see how we did and how other Sharp players in the lobby approached their lineups for this slate. So the two principles we're going to be focusing most heavily on for this video are understanding the odds and understanding the variance and studying the sharks here.

And the tool we're going to be doing using to do this here is SaberSim's contest flashback tool, which takes each contest on DraftKings, the actual lineups that people played, and re simulates the slate 100, 000 times to measure the expected ROI and other useful data points about each lineup and each player that played the contest.

Now this is useful for two reasons, both to help us with those two principles. Remember, this is a game where your average expectation is going to be losing money most nights. Our profit comes from big outlier outcomes, and we'll actually see that here in the data in just a few moments here, but that means it is hard to know if you're actually on the right track.

If you're playing DFS well. If your average expectation is to lose money most nights, so flashback will show us that even if we are, even if we lost money on a given slate or lost money on a given week of DFS, we may or may not still be profitable and we can fix things in our process if we're not showing up as profitable.

Second, it identifies more clearly for us the players that are playing in a way that is going to make the most profit over the long term so we can study from those players. Looking at the top overall lineup on any given night of DFS is not always a good way to learn how to build better DFS lineups.

There's a lot of variance and noise about which lineup ends up finishing in first place. So using flashback makes it easier to identify the most skilled players and figure out what they're doing so we can steal their strategies. So let's pull up flashback here. What I tend to do when I'm pulling up flashback here is I'll just pick a random contest and start looking to see how my contest metrics graded out.

So we'll click into the contest here, go over to the users tab. And I like to do things first. I like to generally look at players that played the same number of lineups I did, or at least as many lineups as I did in the contest, if I didn't max it out. It is easier to have a much higher ROI if you're playing fewer lineups because it's hard to make more and more and more profitable lineups.

So I want to generally compare apples to apples and look at other players that played as many lineups as I did. And then I'll sort by sim ROI, which is really my favorite column here. I think the best overall holistic column, it's going to show you your expected return on investment in this contest. If it played out a hundred thousand times over and over again.

The two other columns that I tend to like to drag over here just as useful data points are the sim median profit. So we'll bring this over and the sim 99th profit. And these are going to show you what your average night, your median average night looks like playing this contest. And also your big nights, your 99th, your top for one percentile outcomes here, when those happen, what they actually look like.

So if I search for my username here, just to figure out where I ended up finishing here, we'll see. That I had an 11 percent ROI with the lineups that I played in this contest, which is a great start that is profitable. Saberson is saying, if you played like this over and over and over again, you'd make about 11 percent profit on every dollar that you end up entering into this contest.

But you can see that my median outcome, my expectation that if I play this most nights, about 50 percent of the time, I'm going to lose 8 or more. And about 50 percent of the time, I'm going to lose 8 or less. It's a little bit weird with the negatives there, but this is my average outcome. So it's a decent way to set expectations.

You can see my 99th percentile outcome here is making 170. Now this is a lower stakes contest, a little bit of a flatter payout structure here. So we're not going to see any 50k wins or anything like that in this contest, but these can set expectations about what these different nights look like here.

You can see on this actual slate here, I ended up making 2 and 56 cents in this contest, which is, you know, it is what it is, but it is actually quite a bit of a better outcome than my average expectation here. So worked out about pretty well, all things considered here. And I just want to check going through each of my contests to see how I graded out by each of these metrics.

So I'll take a look at the minimax next. One thing to keep in mind here. Because once this loads, it is going to bring over your sorting methods here. So I'm going to remove these real quick and instead look at other players that maxed this contest out. We'll again, look at SIM ROI here and see how I graded out in the minimax.

And you can see in this particular contest, I was actually almost perfectly break even. And this is a good opportunity to talk about what your expectations should be about your SIM ROIs. Nobody is going to have a extremely profitable or perfectly profitable SIM ROI in every contest they play on every single night of DFS.

I've checked my results playing MLB DFS pretty much every slate I've played this season, and I've found that I end up grading out with about a 75 to 80 percent of the time I play DFS. I end up with a profitable ROI and looking through what I see from other top users, I think that's frankly on the pretty high end of that.

Remember, at this point, if you have basically a 0 percent ROI in this contest, you're already beating the rake here. We're breaking even. Most of the nights where I have a negative ROI here, it is still beating the rake. So I'm still over the negative 16%, which would be like the average player can't beat the rake player expectation there.

What I'm trying to say is that, although there are certain slates where you're going to show up with a negative ROI, that negative ROI is still beating the rake, and those days are less frequent than your ROI's. where they are in the positive. So even if you have some days that are in the negative mixed in there, you're outweighing those with the days that are showing up in the positive.

And then a lot of times there's days where you're really right. The Sims liked your lineups quite a bit and you're performing quite well. So over the course of the season, you are having a positive ROI. That's really the way you want to think about this is I wouldn't tunnel vision too hard on a singular slate.

Or a singular contest, but instead look at your results over time and get a sense of, are you building profitable portfolios more nights than not? And are your profitable numbers outweighing the negative numbers? If you have a similar ROI in a contest that is under the rake, less than negative 16 percent or negative 10%, That is definitely a cause for concern.

I think that's something that you want to pay a little bit more attention to. If it's something that's slightly negative here, it could very well be that some of the assumptions in the field lineups or other things you took with you into your contest didn't work out that night, and you want to check over time to see how you're doing.

I'll take a look at a few of the other contests here. I want to look at the Four Seamer and see how this one in particular looked, worked out. We can see in this case, so I was actually negative 4 percent ROI in this contest here. So probably my worst overall contest. A lot of that is just variants of what lineups happen to end up in that contest.

I assume there's not a material difference of the way that I played in this contest versus the minimax here, but unfortunate to see that. But again, something that I'm not going to be overly concerned on having followed my results this way over the course of the season. And knowing that this is somewhat of a rare outcome here.

If we take a look at some of these other contests, just to see how things shake out lightly positive in the quarter jukebox, let's take a look at some of the other three maxes and the single entries and just see, so pretty strong performance in the three maxes here. We see some of our higher ROIs, which we will generally expect more often than not, some of your highest ROIs are expected to come in the single entry contests.

Especially the ones at low stakes. This is part of the reason we want to be playing these contests is because they're very soft. There are a lot of bad lineups played into these. So our lineups will tend to grade out pretty well. Overall, basically a breakeven performance in the minimax, slightly negative in the four seamer, but a profitable portfolio overall, which is great, which is good, I think this is.

Means that I'm generally playing overall on the right track. Now, the one other thing I want to mention here, and I think a possible explanation for why these ROIs are a little bit lower than what I've been used to this season, to be quite honest, is we saw in the late swap video, the one that came just before this, we had a pretty impactful late swap that ended up taking place on this slate.

We were very heavy on the Dodgers playing in the last game of the slate. Their lineup was very different than expected because of the Dodgers exposure that we had on this slate. We had a lot of lineup flexibility in our lineups to late swap. We late swap right before the Detroit Tigers and Mariners game here, because we had so much flexibility in our lineup and we use the late swap contest Sims.

What SaberSim did is identified the swaps that would make us the most amount of money that would maximize our ROI from that point on. But I wasn't having a very good slate overall at that point. So what Sabersim is going to do is recognize the fact that my lineups are in general underperforming.

They're certainly not sweating a top 0. 1 percent finish or a top 1 percent finish. So it's going to make a little bit more aggressive changes to my lineups to maximize the remaining ROI I can get. SaberSim's gonna say, for example, if you have this lineup that's doing pretty average heading into these final two games and originally it was a Dodger stack.

Well, it's not gonna win with a Dodger stack at this point. We need an extreme outlier, low ownership performance from one of these teams to end up having this lineup have a chance at winning some decent money. Let's swap that into a Mariners lineup, maybe even leaving some salary on the table, and that is the way to maximize that lineups ROI at that point of what was actually playing out on the slate.

But the post contest Sims that replay the contest from the start, a hundred thousand times over and over again, are going to think that that was a lower ROI decision. So it's a very important thing to note. And I think a subtle point about late swap is when you are late swapping, you are making the best decision that you can for the given.

Ways that the games are playing out in real time, contest flashback may sometimes end up lowering your ROI in the way that it perceives those lineups after the fact, because it's almost starting to slate over from scratch and replaying it differently a hundred thousand times. So it would look at said lineup, like the example I was giving and basically say, why did you play the 3.

5 implied run total Mariners when you could have played the Dodgers in this lineup? It would have been a much better lineup and it would have been for most slates. But for the exact slate that was playing out in real time at around 7. 15 my time last night. It was the right thing to do to swap to the Mariners.

Just something to be aware of. You should still be late swapping. You should still look at contest flashback, but on the slates where you have particularly impactful late swaps, your contest flashbacks may end up. Your sim ROI and contest flashback may look a little bit lower. Those are the main things that I like to look for when I'm just reviewing my.

Results. I'll go through, look at each contest. I played, see how my SIM ROI graded out. I'll take a quick look at SIM medium profit and SIM 99th profit as well, just to get a sense of what my expectations should be for playing these contests here. I am curious on looking at the minimax for example, or the four seamer.

Sometimes these contests can be a little bit more fun to look at the. Sim 99th profit, if you're wondering like where the big nights actually come from, even in a contest where I didn't grade out particularly well, some of my highest profit potential was from the profitable lineups that I was playing into the Foreseemer, like for example, just because I play at a minus 4 percent ROI into this contest here as a overall portfolio, doesn't mean all of my lineups were minus EV.

In fact, some of my lineups here had up to, it looks like a 16 percent ROI played into this contest here. And you can see that Sim 99th profit generally is going to come from these profitable lineups here, having their top 1 percent outcomes, but it's a good idea to just get a sense of, are you on the right track here?

And. What does your overall performance look like and what should your expectations about night to night success look like? Now, if you are following the Saber system thus far, I expect that you'll have pretty similar results that I will. I would hope that your negative ROI nights and contests are going to be under the rake and show up less frequently than your positive ROI nights and contests.

That should show up more frequently and occasionally mix in some probably very high ROI nights depending on your overall stands and takes that you have with your lineups. If you are not having those, it's a good opportunity to lean on the SaberSim coaches to get some extra support. I would attend Office Hours, which is a live stream that takes place every Monday to Saturday on our YouTube channel where you can ask our coach Andrew any questions you have.

Make sure that you're handling things the right way here. And if you're on SaberSim Ultimate, you can email us at support at sabersim. com and either ask for a live build review by letting us just know what slate you played, and a coach will review your actual build from SaberSim and call out any mistakes that they're seeing, or you can just ask a question and a coach will send you a personalized video response here to give you some extra support.

I've seen some players reach out before with negative sim ROIs that are struggling that feel like they're doing everything right. And all it takes is a 15 minute build review for one of our coaches to look at your build and say, Oh, these few things are wrong. This is where we can shore this up. And you may immediately be able to turn that around.

And if you're not on Ultimate, the Office Hours is also a great resource to get that opportunity just in a live stream format. to ask any questions you have and get your answer, get your questions answered there. But, of course, all of us DFS players are constantly trying to get better, myself included, even when I see profitable sim ROI numbers or have a few good slates in a row.

I'm always trying to get better and better and better. The best way to do that is to look at the players who are the most successful in the large field flagship tournaments. So that is the second part of this here and something that I spend frankly a little bit more time on most times I'm looking at flashback here is I will go ahead and look into the flagship GPP, the big 150 max and pull this up.

And this time I'm looking to see who the top. ROI players were that maxed this contest out, um, which players that played this contest had the highest SIM ROI here, the highest expectation of profit, even if they lost money on the slate here. So there are some names that you will tend to see show up over and over and over again here.

OCD, OBV shows up all the time. We see Royal Pain up here a lot. Pickled Bird. BB Mon shows up quite a bit. It looks like Mr. Shady Advice may have taken this slate off, but I tend to see his name up there as well quite a bit. And the main thing that I like to do here is to go in and take a look at what the top players by SimROI that 150 maxed this contest, how they approached their lineups, and in particular, answer any questions that I had as I was building my lineups to see if Maybe there's something I can learn from there.

We can't go back in time and replay the slate and just copy exactly what these players did, but we can pick up on patterns and take them with us into future contests. So if we look over here at the top 150 maxers lineups here. One of the questions that I did have as I was building here is I ended up with a pretty big bag of Dodger stacks heading into this slate.

I had about 40 percent Dodgers here. It was a little bit of a weaker lineup that they played. Even the one we were expecting at lock before we late swapped was expected to be a little bit of a weaker lineup. It's a team I had a lot of exposure to. And even as I was building and diversifying with Minioniques and things like that, I still felt like I might be a little bit exposed.

We do see that at least the top 150 maxer here was off of that stack a little bit more than I was here. And frankly, just a little bit more diversified overall. And if we start to look at some of the other players here, we see a similar thing. So a little bit lower on these Dodger stacks overall. Now this player was pretty heavily focused on the Milwaukee stack.

So a little bit higher exposure there. I'm going to check a couple others here. and see how some of these other royal pain is somebody that I always see up at the top here. So a little bit higher Dodgers exposures, but quite a bit flatter exposures overall. So it might be something to take with me here is to remember, even if I am following my diversification processes I've outlined before, and things like that.

If I am getting a very large exposure to one team, it might even be a good reason to bring down that exposure a little bit here. I don't think it was bad process to have a lot of Dodgers bats on last night's slate. Uh, but seeing how some of these exposures look, I'm wondering if maybe I should have brought that exposure down a little bit more.

Now, there's not going to be any single skeleton key that you just discover looking at one player's lineups for one player's slate. And I think it's important to try to not. over tunnel vision on what you're seeing in just a small sample here. But over the course of the season, looking at what other players are doing, you will start to pick up on patterns.

Some of the takes that I have about what stacks I like to use in baseball DFS, for example, have come from and also my approach to diversification in general, I've come from looking at other players over the course of the season and seeing how they're approaching their lineups. That's why to play generally pretty diversified.

And that's why I do like to edit the stack types that I'm using on these slates. I've picked up all kinds of these different things across a variety of different sports using Contest Flashback to dig into what these other players are using, and it can be really a useful way of reviewing and improving the quality of your play overall.

The one other note I'll make here is it can be a little bit hard the next day to pull the slate back up and try to figure out what you're looking for. So as you're actually building your lineups, if something sticks out to you and you're unsure, Should I do this? If you're getting a low owned stack popping into your builds from the sims even after diversifying and you're thinking, I don't know if I should play this team here, make a note of that.

Or, for example, last night I made a mental note of, That's a lot of Dodger stacks, and I made a note of that. Just the things that are popping out to you as you're going through the process of building your lineups if you're unsure of how you want to handle a particular situation, make a note of it and check back the next day.

Again, over time, you'll start to get a sense of, Oh, I remember the last time I was unsure of a situation that felt like this. A lot of the sharp players went this way, so I'm gonna go that way this time. It'll start to build up a sense of intuition about how you can take certain stands in certain spots that resemble situations you've seen before in your DFS play.

Now, I really can't understate how valuable flashback is here as a tool to both understand your odds and your variance, and also steal strategies from the best players and see how they're approaching particular spots. So if you're serious about improving your DFS play, this is a very important part of the process and something that I definitely make some time for the next day or the day after that, after you've played DFS.

It's a regular part of my process. I actually get pretty excited to look at contest flashback the next day every single time I'm playing and I highly recommend it as one of the main principles that you need to be successful in DFS. Now that is going to wrap up the Saber system here. If you've stayed with me thus far I really appreciate you watching this and if you have any questions you can reach out to us at support at sabersim.

com. This is a new type of content that we're experimenting with here as this full course system. The plan is for us to re record these more practical episodes frequently for new sports, new product updates on SaberSim, new model updates, and constantly iterate on what we're working on here, delivering a fully practical guide to using SaberSim to beat DFS.

So if you enjoyed this, let me know. If you have any other questions, send us an email. In the meantime, thank you for watching, and good luck.

‍