Best Three Trading Indicators. Learn This Powerful Three

Against Buying Low: A Meditation On Our Favorite Fantasy Tactic, Which Might Not Actually Work

Warning: Novel-length post ahead.
It’s that time of the season when this forum—along with much of the fantasy basketball punditry across the web—is fixated on buying low. It seems like the majority of threads here, and columns across the fantasy basketball web, are focused on identifying good buy low candidates.
Which, to me, begged the question: does buying low actually work? And the answer was: I had no fucking idea. Sure, there was a wealth of anecdotal evidence, but that could lead you in very different directions, depending upon whether you had bought low last year on Danny Green, or on Chris Paul. There was no metric of how effective a strategy it is on the whole.
So I set out to create one.
There’s a lot of uncertainty involved in buying low on players, and so by necessity we should be making probabilistic predictions: talking in terms of the odds that something will happen, rather than a binary will-it-or-won’t-it position. Anyone who says “Player X will certainly bounce back from their slow start” or “Player Y certainly won’t” is talking out of their ass. The truth is, we can’t say for sure: NBA statistics are the product of a complex system, not quite as complex as, say, the weather, but complex nonetheless. I can state with a pretty high degree of confidence that Kawhi Leonard will post top 10 value over the rest of the season (much like a meteorologist can estimate with pretty high confidence what the high temperature will be tomorrow). But ask a meteorologist what the high temperature will be on January 20th, or ask me whether Paul Millsap will post top 25 value the rest of the season, and now neither of us is quite so confident. The meteorologist would be hard pressed to do better than just guessing the average January 20th temperature from the last 20 years. And I’d probably be best off trying to figure out guys similar to Paul Millsap who have started slow, and asking what percentage of them ended up posting top 25 value.
Now, of course I am not suggesting that a statistical model could give you all of the information you need to determine whether it’s a good idea to buy low on someone. Things like team situation, minutes, injury history, etc. matter of course, and there’s no easy way to build a statistical model to account for them. But it’s still essential that we have some sort of model to give us an overall understanding of how effective buying low is. When we’re thinking about buying low, we tend to start with the “inside view” (whether a guy’s coach likes him; whether he’s spending extra time working on his FT shooting; whether he’s in a contract year), but in reality, we should be looking first at the “outside view:” what % of all buy low candidates end up getting their shit together? That number should be our jumping-off point, and then, when we have it, we can adjust it for situation-specific info like their minutes, their team situation, etc. If we start by looking at the inside view, we can fool ourselves into thinking: “Everything’s set up for this guy to succeed; there’s probably a 90% chance he bounces back this year.” When, in reality, that’s probably a pretty terrible estimate, and the outside view shows us why.
[A note on the data (feel free to skip this): I have taken players who have started slow over the first 32 days of the previous 3 seasons (2013-14, 2014-15, and 2015-16). These are “struggling” players. In each year, I have limited the data set to players who were in the top 75 in final average value the previous year, per bbm, and I define “struggling” players as those who are underperforming their previous year’s ranking by at least (5 x round #). So if a player finished 32nd the year before, they would need to be at 47th (32 + [5 x 3]) or lower over the first 32 days of the following season in order to qualify as “struggling.” Of course, these distinctions are, to an extent, arbitrary, and you could slice the data somewhat differently (though I think the general principle would still hold). My rationale for cutting off the pool at top 75: players below that level generally aren’t “buy low” candidates, because their owners tend to cut them if they’re really struggling early in the season. As for determining what qualifies someone as “struggling,” I wanted to capture the fact that a 1st round pick who slides 10 spots is a much bigger disaster than a 6th round pick who slides 10 spots. I don’t think the method I came up with is perfect, but it does at least have a reasonably even distribution of “struggling” players in each round. Also, I tossed out anyone who played fewer than 50 games in their prior full season, or fewer than 10 games over their “struggling” month. I wanted this to reflect guys who are basically healthy; buying low on injured players is a whole other can of worms.]
There are 78 “struggling” starts over these 3 seasons. (And some guys lay claim to multiple slow starts; looking at you, Serge Ibaka.) In looking at end results, I have grouped players into 5 categories. Category 1 encompasses players who, for the remainder of the season, actually outperform the value they put up the previous year. So if a guy was ranked 30th in 2014-15, and then slides to 60th in the first month of 2015-16, but then posts 25th-ranked value for the rest of 2015-16, he’s in Category 1. A player in Category 2 did not match his value from the previous year, but came close enough to it over the rest of the season that he wouldn’t qualify as “struggling,” per the definition above. Players in Category 3 qualified as “not struggling” under a less stringent definition: 10 x round #, instead of 5 x round #. Category 4 encompasses players who improved upon their terrible start, but didn’t play well enough to qualify for Categories 1, 2, or 3. So a player who ranked 40th the year before, and then 120th over the first month of the season, and then 100th over the rest of the season, would fall into Category 4. Players in Category 5 went on to perform even worse for the rest of the season than they did in the first month.
In more basic terms, you can think of it like this: Buying low on a Category 1 player is an Excellent Decision, buying low on a Category 2 player a Good Decision, buying low on a Category 3 player a Neutral Decision, buying low on a Category 4 player a Bad Decision, and buying low on a Category 5 player is a Terrible Decision. So generally speaking, in a reasonably competitive league, in order for a buy low trade to actually help your team, you’ll need the guy you receive to fall into either Category 1 or Category 2.
[Aside: There’s no way for me to know how intelligent the other managers in your league are; of course, if you can get a struggling guy who was drafted in the 2nd round for waiver wire fodder, then go forth (and you don’t need this guide). But in my experience it’s usually hard to buy really low on medium-to-high draft picks. You can cite the sunk cost fallacy all you want, but on some level, owners’ aversion to selling low makes sense; your objective should always be to win your league, and if you sell a guy that you took in the 2nd round for someone whose ceiling is 6th round value, you may have improved the floor of your overall team, but you’ve almost certainly made yourself less likely to come in 1st than if you just held onto your 2nd round pick and prayed.]
So let’s take a look at the data. What percentage of struggling guys end up falling into each of our 5 categories?
The breakdown isn’t very encouraging:
Category # of Players
1 15
2 20
3 10
4 11
5 22
http://imgur.com/a/8yCKy
If you buy low on a random player, the single most likely outcome of these 5 is that he’s going to go on to play even worse than the month of games that made him a buy low candidate in the first place. There’s a 42.3% chance that buying low on him is going to be a Bad Decision or a Terrible Decision, and a 44.8% chance, roughly comparable, that it’s going to be a Good Decision or an Excellent Decision. And given the sort of value you usually have to give up to buy low, it’s likely that the Category 4 and 5 players are going to hurt you more than the Category 1 and 2 players are going to help you. (I.e., it’s unusual for someone who is top 75 to begin with, and then sucks for a month, to suddenly get it together and start playing significantly better than their baseline ability. But it’s not that unusual for someone to suck for a month, and then continue sucking for the next 4 months.)
“Okay,” you might say, “Buying low in aggregate isn’t an amazing idea, but given my fantasy basketball knowledge, I can determine which players are going to be Category 1 and Category 2 guys, rather than just buying low at random.”
This was basically what I believed about myself. So I thought about which factors might lead me to think a single particular player was a good “buy low” candidate. The first thing that came to mind was the round they would have been picked in, based off of value the previous year. Surely 1st and 2nd round picks are safer investments, more likely to sniff 1st or 2nd round value despite their slow starts than a 6th round pick would be likely to approach 6th round value if he starts out badly. Right?
Well, basically, no.
If you look at the breakdown, higher-round slow starters are just as likely to flame out as their lower-round counterparts. And they aren’t any more likely to exceed their value from the previous year, either.
Round Category 1 Category 2 Category 3 Category 4 Category 5
1 2 6 0 3 4
2 3 6 1 2 4
3 3 2 4 3 5
4 2 2 2 3 4
5 3 2 1 0 1
6-7 2 2 2 0 4
Round Category 1 Category 2 Category 3 Category 4 Category 5
1-3 8 14 5 8 13
4-7 7 6 5 3 9
Then I scrutinized the list of high-ranking players, and wondered if there was a better indicator than round value to demonstrate someone’s consistency or safety as a pick. After all, Danny Green posted 2nd round value in 2014-15, but no one was drafting him in the 2nd round in 2015-16. So how about sorting guys by their usage rates? Usage is a pretty solid indicator of how involved a player is in his team’s offense. It stands to reason that someone like LeBron, who has a sky-high usage every year, is a safer selection, relative to where he’s drafted, than someone like Danny Green, who touches the ball less and thus is more vulnerable if his team’s tactics evolve (like, say, to accommodate the arrival of LaMarcus Aldridge).
So I sorted the players in my data set into 3 categories: High Usage (someone with a usage rate above 26.5 in the full previous season); Medium Usage (someone with a usage rate between 20 and 26.5 in the full previous season); and Low Usage (someone with a usage rate under 20 in the full previous season).
Surprisingly, the correlation isn’t especially strong here either. It’s true that on the whole, high usage players are somewhat safer bets, and are less likely to end up in true disaster territory (Category 5) than low usage players, with the middle-usage guys falling in the middle. But it’s not a hugely strong correlation, and it’s also true that low usage players are more likely to fall into Category 1 than high usage players are.
Usage Category 1 Category 2 Category 3 Category 4 Category 5
High 5 9 1 5 4
Medium 2 9 6 2 8
Low 8 2 3 4 10
http://imgur.com/a/9bTFv http://imgur.com/a/ltLJT http://imgur.com/a/XS6IE
Next, I wondered whether position might be the better metric to look at. Sure, some of the benefits and drawbacks of position are captured by usage (with Cs and SGs tending to have lower usage rates than PGs and SFs), but perhaps there were other pieces of the puzzle that usage didn’t capture. Maybe centers, due to higher risk of injury and a more dramatic dropoff late in their careers, were at greater risk of not recovering from a slow start. Maybe point guards were more insulated due to their central role in a team’s offense.
But, running the numbers, there wasn’t much to glean here either. Perhaps big men are slightly riskier buy low candidates, but the upside seems to be greater as well.
Position Category 1 Category 2 Category 3 Category 4 Category 5
PG 2 5 2 3 3
SG 1 5 1 2 5
SF 4 3 2 2 3
PF 3 3 2 1 5
C 4 4 3 3 6
Then I looked at the fantasy category in which each buy low candidate was struggling the most. Maybe guys who start the year in a shooting slump are more likely to bounce back than guys who see a reduction in rebounds, for instance. I had expected that FG% would be the most common category here, but I was shocked by just how common it was. 34 out of 78 players (43.6% of them) had a greater dropoff in FG% than in any other category. The other very popular category was steals, which was the biggest problem for 13/78 players (16.7%).
Based on the data I’ve got here, there’s no category that seems to clearly suggest whether a player is a good buy low candidate. Even dropoffs in points and rebounds, which you would think would be indicators that a player has entered a less favorable team situation, don’t tell us much about how they’re likely to perform moving forward (though the sample is, admittedly, limited). This suggests that even players whose situations change dramatically usually find ways to make themselves useful from a fantasy perspective, whether it’s improving their efficiency (FG% and TOs) or focusing more on defensive stats. A dropoff in 3PT is probably the most concerning based on this data, but with just 6 players in that category, I’d be wary of drawing too firm a conclusion from that. On the whole, the category a player struggles in isn’t especially helpful in projecting their performance moving forward.
Fantasy Cat Category 1 Category 2 Category 3 Category 4 Category 5
3PT 1 1 0 1 3
AST 1 0 0 0 0
BLK 4 0 1 0 1
FG 6 11 3 4 10
FT 2 0 1 3 0
PTS 0 4 1 1 1
REB 0 1 1 1 0
STL 1 3 3 1 5
TO 0 0 0 0 2
So what about age? Finally, here, a pattern begins to emerge:
Age Category 1 Category 2 Category 3 Category 4 Category 5
24 and under 2 4 0 2 0
25 5 6 0 1 3
26-28 2 1 4 3 7
29-30 5 6 3 3 4
31 and above 1 3 3 2 8
Age Category 1 Category 2 Category 3 Category 4 Category 5
25 and under 7 10 0 3 3
26 and older 8 10 10 8 19
http://imgur.com/a/kZj1u
At first glance, this surprised me. Certainly, I expected there to be a significant dropoff as players aged into their 30’s, as this data includes guys from the Garnett/Pierce/Nowitzki generation as they gradually slipped from reliable early round players, to middle-rounders, to scrubs. So that piece of the aging curve (with the brutal track record for players 31+) is pretty much exactly what I expected.
It’s the younger half of the aging curve that initially confused me. I think the traditional understanding is that athletes in many sports, including basketball, tend to peak statistically around the age of 26-28. So I had thought that we would see the best buy low options in that age range. Instead, it skews significantly younger, with age 25 appearing to be the cutoff: guys 25 and younger are, on the whole, much stronger buy low candidates, and then once they hit 26, it’s no longer a good idea to buy low on them.
But the more I thought about it, the more this made sense. After all, when we talk about guys peaking in the 26-28 range, we’re talking about their absolute performance, whereas in fantasy we’re more concerned with their performance relative to the previous year, and hence relative to where they were likely to be drafted. And players almost always make the biggest gains in fantasy value between the ages of 22 and 25. Now, of course, some of this expectation of improvement is baked into guys’ average draft positions; players like Towns and Porzingis are going to be drafted above where they finished the previous year on the expectation that they will develop and improve. But it seems like this assumption of improvement often goes out the window when a young guy starts slow; people are often willing to cut the cord on underperforming young guys, whether due to a fear of the “sophomore slump” or a suspicion that, given their limited NBA track records, they may just not be good enough yet to justify their high ADP (see, for example, all of the people panicking about Towns’s somewhat slow start).
In my experience, people seem to be more willing to unload an underperforming young player at a reasonable price, than they would be willing to unload an underperforming 30-year-old with a solid fantasy track record of many years. But they shouldn’t be: buying low on young players is not only higher-upside, it’s also, surprisingly, a safer bet than buying low on veterans, or even buying low on guys in the 26-28 range, which I would have thought to be the safest age range of all.
So if there’s one thing you should look at when you consider buying low, it’s the dude’s birthday.
Now, is there an appropriate time for buying low on guys older than 25? Of course. It’s important not to lose sight of the fact that (for me, at least, and I suspect for many other players) fantasy sports are an all-or-nothing exercise, e.g., you’d rather give yourself a chance of coming in 1st and a chance of coming in 8th, instead of locking in a 4th place finish. So if your team is struggling, it makes sense to pursue volatile, high-risk-high-reward strategies. And that’s exactly what buying low is.
But on the whole, buying low is not an efficient way of generating value. Trades rarely are. The most reliable way to generate value is to draft well; then, trades should be used primarily to solidify punting strategies, not to generate value. Sure, occasionally you’ll strike it hot with a random buy low candidate, but overall, it’s not a reliable way to improve one’s team. Yes, if your team is already bad, you might as well leverage a high-risk strategy and give it a shot, but “buying low” is often touted as THE essential fantasy basketball strategy, and there’s not a lot of evidence to support that.
submitted by Rodekio to fantasybball [link] [comments]

Binary Options Trading Strategies - YouTube BINARY OPTIONS - Trading with Binary Option Strategies - guaranteed profit - IQ Option Trading Binary Options Trading Strategy 2014

Start with CWOption‘s $50 No Deposit Binary Bonus to take an adventure of live binary options trading in a risk-free way. Join the web-based binary options […] Get Bonus CWOption Review. 31 December, 2014. Latest Bonus Update Weekly Special. 115% BONUS ON EACH DEPOSIT – TTS Markets. July 26, 2020. Trade2Win Draw, $10k Fund every Week Foreign exchange trading pros and cons.. This software is in .exe format and you will learn advanced Strategies so you don't buy any foreign exchange trading courses, or attend classes or seminars, read Pdf ebooks, video tutorials, or download zip or rar, forex Trading forums. Share this video on twitter and facebook.. On Friday, October 3, 2014, the US Unemployment Rate as well as the Nonfarm Payroll report will be released at 8:30 AM. There are many Forex pairs that you can trade Iron... masterforex v options binary options minute strategy guides torrent binary-options-trading-2014-15-tips-and-facts-for-a-winning-strategy-5 minute binary options trading strategies forum strategy by forex brokers minute jya binary translation hours is binary Binary options trading signals reviews top 10 Options guru and veteran floor trader, Frank Fahey, leads off discussing de-funding the police, a violent weekend in Chicago and much more. John Flanagan continues the conversation about shootings in the city before discussing the economic impact of a government-mandated shutdown and importance of removing statues and monuments.

[index] [27599] [16546] [201] [30097] [2582] [9535] [9051] [20016] [5782] [8940]

Binary Options Trading Strategies - YouTube

All about binary options trading. Learn to trade binary options correctly. We’ll talk about strategies, indicators and technical analysis in your language. BINARY OPTIONS - Trading with Binary Option Strategies - guaranteed profit - IQ Option Trading We make videos using this software bot which aims to make it easier for you to trade, because to use ... This channel created for all Binary Options Traders, Follow this channel to learn Live Binary Options trading Strategies, Avoide costly mistakes and learn ab... Binary Options Trading Strategies For Beginners Binary Options Strategy Video Tutorial 2014 - Auto Binary Signals, a piece of software designed by Roger Pierce, who claims it made him his ... trading binary com options strategies and tactics - https://bit.ly/3faoQYT Strategy is a key element of long term successful binary options trading. The best...

Flag Counter