If there is one stat you hear thrown out more than any other around the Florida basketball program, it’s this:
Assist to turnover ratio.
This stat is, well, exactly what it sounds like. Take the number of assists and divide it by the number of turnovers. For obvious reasons, you want that number as high above 1 as possible. It’s a number people have used particularly when evaluating point guards as a way of quantifying the responsibility of guards and many coaches, media, and fans are all over it.
When White was first hired to lead the Gators he initially used the stat to discuss his desire to take care of the basketball, something the Gators have done particularly well in his tenure.
White Kasey Hill graduated and Chris Chiozza took the reins at point guard much of White’s compliments towards him revolved around his assist to turnover ratio. Before Chiozza’s senior season, White openly challenged him to lead the country in assist to turnover ratio (he didn’t).
Now, assist to turnover ratio is back with White as well as his trio of point guards in Tyree Appleby, Tre Mann, and Ques Glover all with those words on their tongues. White has had to field multiple questions about who will start or play the majority of minutes and point guard and on multiple occasions answered that assist to turnover ratio would be the deciding factor. When Glover asked about what his focus was entering the 2020-21 season he said it was “all about” assist to turnover ratio. Mann and Appleby are also plenty aware of how much assist to turnover ratio is being discussed and have both mentioned their focus on the number.
For the first time since Kasey Hill and Chris Chiozza were on the roster there is going to be major competition at the point guard position. When Hill graduated it was unquestionably Chiozza with the starting job without a big push from 3-star freshman Eric Hester. Then Michael Okauru came to town, and while he provided some decent minutes there was no question about the fact Chiozza was going to take every second on the court when he wasn’t fatigued or in foul trouble. Fast forward another year and Andrew Nembhard came in as a 5-star ready to take the starting role and with Okauru not making a jump in his sophomore season Nembhard was the unquestioned owner of the position. Tre Mann and Ques Glover came on campus last year with a chance to push Nembhard for minutes but neither freshman was able to earn the trust of the coaching staff, and Nembhard continued to control the lead guard role.
Now, the point guard competition has seriously ramped up between Tyree Appleby, Tre Mann, and Ques Glover. Each of them wants a chance to start and play big minutes, and each of them bring some different skills to the table. Appleby is the best passer out of the group, and also has shown tremendous shooting touch. However, he’s a bit undersized and isn’t a great defender, or at least he wasn’t at Cleveland State. Mann brings outstanding size and athleticism to the position as well as loads of scoring talent. Unfortunately, he wasn’t able to realize that talent as a freshman and while all reports from camp are that he’s scoring the ball at a torrent rate, he’s not a great passer or distributor. Glover was another player who shot the lights out in college but struggled in his freshman season, and he might end up being the best ball pressure defender out of the three with his lateral quickness.
Entering the offseason it was largely expected that Appleby, a redshirt junior this year, would be the leader in the clubhouse. As a fourth year player he has plenty of experience and had two productive years at Cleveland State where he proved his ability as a high level offensive creator. However, White has said that Mann is probably the most likely to start after coming into camp vastly improved from his sophomore season. In the same press conference he also said that there were days where Glover was the best of the three in practice, insinuating that he’ll have the opportunity to start if his practice effectiveness continues.
With the competition for point guard minutes so close, minutes could very well come down to assist to turnover ratio like White has said it will.
That begs the question–is that a good way to evaluate point guards?
This is going to fly in the face of a lot of basketball people who love the stat but really, no, it’s not. Assist to turnover ratio is a simple statistic that gives far less information than a lot of people think, and is riddled with multiple problems. Here is why assist to turnover ratio shouldn’t be used as a way to evaluate point guards against one another.
First, let’s talk turnovers. An assist is a passing stat (obviously), and assist to turnover ratio treats turnovers as a failed pass. However, that’s not the case. There are a ton of ways to turn the ball over such as travels, ball handling mistakes, and committing charges. Looking at a point guard’s overall turnover numbers isn’t an accurate way of looking at their passing mistakes, because the turnover stat is a noisy one with multiple inputs and it’s not entirely fair to compare that next to assists which is a very linear, single-input statistic.
Isolating passing turnovers would be a much more accurate comparison if assist to turnover ratio had to be used. Passing turnovers is a stat available on the advanced NBA statistics page, but you don’t see it kept anywhere for college statistics. Keeping passing turnovers as an internal stat within the program wouldn’t take a lot of time or effort and would be an improvement over simply looking at the overall turnover number.
Another problem with assist to turnover ratio is that it inherently makes a one-to-one comparison between the value of an assist and the penalty of a turnover. From a basketball analytic point of view, that simply isn’t the case. Let’s show you why.
An assist is inherently worth 2 or 3 points on the board. If a player has 5 assists, it means those assists are worth at least 10 points and as much as 15, depending on what kind of shot the assist led to. However, 5 turnovers does not equal 10-15 points lost, as the assist to turnover ratio would imply. A turnover is a loss of possession, which hurts, but it’s not automatically putting points on the board for your opponent.
Let’s pretend every turnover resulted in a fast break for an opponent which isn’t always the case, but sometimes it is. Transitional plays always yield better results than halfcourt plays, so fast breaks off turnovers are high-value possessions. Even if that’s the case, the predicted points on a fast break in college basketball were 1.1 points per possession last season. Therefore, a 1.1 point penalty is a fair way of saying what a turnover is worth, and you could even argue that it’s less considering not every turnover leads to a transition opportunity.
Simply put, the value of an assist is 2 or 3 points.
The penalty of a turnover is around 1.1 points.
They are not equal, and assist to turnover ratio treating them like they are makes for a misleading stat.
Let’s look at a hypothetical situation.
Player A has 3 assists and 1 turnover.
Player B has 6 assists and 4 turnovers.
On the surface, many people would look at Player A and think he had the better, more responsible game. He had an assist to turnover ratio of 3, one that coaches who love the stat would be extremely happy with. Player B, on the other hand, had an assist to turnover ratio of 1.5 which would make a coach that likes the stat not pleased.
However, let’s take a deeper look into the numbers.
The average assist in college basketball led to 2.4 points last season (there were more assisted 2 point baskets than 3 point baskets). We can use that number as a fairly accurate look at what an assist is worth. Now, we can more accurately quantify what these players accomplished for their teams.
Player A had:
3 assists equalling +7.2 points for his team.
1 turnover equalling -1.1 points for his team.
His net impact of assist and turnovers was +6.1.
Player B had:
6 assists equalling +1 points for his team.
4 turnovers equalling -4.4 points for his team.
His net impact of assist and turnovers was +10.
At first glance, anyone who subscribes to belief in assist to turnover ratio would have thought Player A had a much better impact on his team. However, when you actually look at the points created and the points given up, Player B had a drastically better game.
That’s because of this, a sentence that sums up why assist to turnover ratio is inaccurate.
The creation of assists is far more valuable to a team than the penalty of turnovers is detrimental.
Basketball is all about creating advantage situations which moves defenders and frees up players for open shots. The ability to create open shots, whether for the individual or for a teammate, is the most valuable skill in all of basketball. Players who create a large number of high-quality shots should not be punished for having a large number of turnovers, even though old-school coaching philosophy is still to look at turnovers like they’re as bad as sin.
A perfect example of this can be found on Florida’s very roster. Tyree Appleby averaged 3.4 turnovers per game at Cleveland State, something that had certain people disgusted when he came to Florida. That’s because people are conditioned to see that many turnovers and be appalled. However, looking at raw turnover counts isn’t a great way of evaluating a player’s ball security. You see, Appleby was one of the highest usage players in the country at Cleveland State who had the ball in his hands a ton, and when a player has the ball in his hands for most of the game he’s going to turn the ball over more than a shooter who sits in the corner and barely touches the ball. A more accurate stat is turnover percentage, the percentage of plays the player turns the ball over. Tyree Appleby’s turnover percentage was actually better than Andrew Nembhard. Additionally, Appleby had the 11th highest assist rate (percentage of plays that end in an assist) in the country.
Appleby averaged 5.6 assists to his 3.4 turnovers, an assist to turnover ratio of 1.8. That number wouldn’t be looked on extremely positively by the assist to turnover ratio crowd, but what about the actual points produced minus points surrendered?
5.6 Assists x 2.4 PPP = 13.4 Points Produced
3.4 Turnovers x 1.1 PPP = 3.7 Points Given Up
13.4 Points Produced – 3.7 Points Given Up = 9.7 Points Of Value Added
9.7 points a game for Appleby added in addition to the fact he scored 17 points per game on his own. His ability to create shots for others in volume easily makes up for the turnovers he gave up, and his net value as a high assist/high turnover player is much more valuable to a team than a player who has a better assist to turnover ratio but accomplishes it by having much smaller numbers of assists.
Of course, in the assist to turnover ratio conversation it’s not just turnovers that are a problematic stat. Assists are another number that aren’t entirely accurate as it relates to a player’s ability to pass the ball and create offense for others. That’s because a passer can do his job precisely–collapse a defense, find a wide open man…and that player can miss the shot. The passer did his job perfectly, but doesn’t get rewarded at all.
There is a way to properly evaluate players as passers and assist creators, and that’s to look at their shot creation ability without even factoring in if the shot ends up going in or not. That’s because it shouldn’t matter if a player makes a shot or blows a layup, if the passer created a quality shot they should get rewarded. This is referred to as a “shot quality assist,” data best found from the tool Shot Quality. This powerful analytics tool credits players for creating high quality shots that are likely to go in, something that credits a player for creating a wide open shot or layup that’s missed. As well, if a passer creates a high-percentage shot and the player gets fouled, he’ll get credit with a shot quality assist. On the other hand, if a player passes the ball to a teammate who takes a bad, low-quality shot and it happens to go in–the passer won’t get credit for a shot quality assist.
For example, Nembhard finished last season averaging 5.5 points per game. However, he had 6.6 shot quality assists. This shows some bad luck, as he created open, high quality attempts for teammates that didn’t always fall.
Finding less problematic ways to evaluate assists and turnovers make for a much better way of evaluating players than the raw assist to turnover number. Ultimately, you just wouldn’t want to see Florida’s staff making decisions based on a stat that was flawed in multiple ways.
A lot of numbers have been used in this article and while they have all been used to tear down the assist to turnover ratio stat it doesn’t account for one intangible element.
Even if the stat is bad, if it motivates players to play better on the whole, that’s a positive.
All of Florida’s point guards have talked about assist to turnover ratio and even if it’s a critically flawed stat, if it causes them to work on their passing and create more assist opportunities for their teammates, it’s a positive. The fear would be that these players choose to play far too conservatively, fearing the consequences of turnovers and instead choosing to not try to create enough assists for their teammates.
Earlier in the article I mentioned that Florida has been stellar at taking care of the ball under Mike White, averaging a 68th finish nationally in turnover ratio in the country during that time.
However, in that same period they have averaged 240th in the country in assists generated.
Sure, the great turnover percentage numbers are great, but they are coming at the expense of low assists, and low assists means a low number of quality shots being generated by the pass.
As long as assist to turnover ratio is the stat that is held in higher regard than any other, it’s likely this trend is going to continue. That trend continuing would make some people happy, but until the Gators start generating more assist opportunities and more quality shots without fear of committing more turnovers their offense is going to have a bit of a ceiling.
For Florida’s point guards, assist to turnover ratio might be the stat of the day but hopefully as the season progresses it will be points added and points created as well as proper understanding of what the penalty of a turnover is more than anything else.