FAQFAQ Pool and Billiards Balls

... physical characteristics of pool balls.

Dr. Dave's answers to frequently-asked questions (FAQs), mostly from the AZB discussion forum



cleaning and surface treatments

What effects do different ball cleaners and polishers have on the reaction of the balls?

The following video shows the results of an experiment showing how different surface treatments affect throw and cling/skid/kick:

For more information, see "Throw Follow-up: Part I: Cling" (BD, July, 2014) and "Throw Follow-up: Part II: More Results" (BD, August, 2014).

Chalk brand can also have an effect on cling/skid/kick. For more info, see the chalk comparison resource page.

See also: Silicone spray effects.


Why shouldn't we treat pool balls with whichever wax reduces throw as much as possible?

If there were no throw, shot making would be easier because you could aim every shot, regardless of angle, speed, and spin, to hit at the ideal ghost-ball position along the "line of centers." However, throw shots and spin-transfer shots would no longer be possible. Also, as the wax wears off with use, the conditions could change significantly. Also, if everybody didn't use the same wax and clean balls frequently, conditions could be very different from one place to another, from one day to the next, and from one ball to the next.


What ball cleaning product is generally recommended?

Aramith Ball Cleaner is generally recommended.


Why does the amount of throw increase more for the 45° cut as compared to the 30° cut, where throw is usually maximum?

Maximum cut-induced throw (CIT) under typical conditions occurs close to a 1/2-ball hit, or a 30° cut angle (see maximum throw). At a slow 1/2-ball hit, under typical conditions, the cue ball (CB) is sliding on the object ball (OB) during the entire contact period until near ball separation, where the CB and OB just begin to "gear" together with no more sliding. This allows the CB to create the largest amount of throw possible for that angle. With a 45° cut, under typical conditions, the CB is sliding on the OB during the entire contact period so the balls don't gear together during collision. However, with cling, the amount of friction is larger and the CB is able to "gear" with the OB during contact for cut angles larger than 30°, producing more throw. The extra friction doesn't increase throw at small cut angles because the balls are already "gearing" together under typical conditions, already producing the most throw possible. Once the balls gear together, no more throw is possible. With larger cut angles, an increase in friction increases the amount of spin transfer (leading to possible gearing) and throw. If gearing is already occurring, increasing friction further has no additional effect.

For more information, see the plots on the last two pages (pp. 16-17) in TP A.14 - The effects of cut angle, speed, and spin on object ball throw.


from Dan White (in AZB post):

I have been using Novus 1 for cleaning balls in my ball machine for the last couple of years, based on recommendations I saw in AZB. I came across Dr. Dave's video on throw with various ball cleaners, and it got me to wondering how Novus 1 matched up (Dave didn't test Novus 1). The other cleaner I have is the cream from Aramith, which you would assume is the best thing to use since they make the balls. I never liked using Aramith because it is thick and flaky and seems like it would gum up a ball machine quickly. Also, it is a mess to apply. Novus, on the other hand, can be diluted and easily sprayed on the balls in the machine. After doing this experiment, I am a little up in the air as to what the "right" cleaner is. It would seem that consistency is important, which in my experiment favors Aramith. What cleaners are used in tournaments? Is there a standard? What is used in the 14.1 events, where slipperiness can have a great impact on how well balls open?


Originally, I just wanted to see where Novus fit in compared to Dave's results. However, I began to see something odd, so I decided to take Dave's experiment one step further. For each ball in my experiment, I shot it over and over to see if the amount of throw changed over time. Typically I shot each ball at least 20 times. I did find some interesting changes (summarized below).

First, take a look at Dave's video to see the experimental set up.


For those of you who want the short version, here are my results. Note that I hit the balls a little harder than Dave did so that makes sense that I got slightly less throw with Aramith than Dave did:

1. When I cleaned the balls with Aramith (and buffed them in the ball machine) I found that throw started out at 2 inches, and within 8 to 10 trials, the throw increased to 4 inches, and remained there.

2. When I cleaned the balls with Novus 1 (and buffed them in the ball machine) I found that throw started out at just under 2 inches and took about 20 trials to reach 4 inches of throw. I also found that the amount of throw was inconsistent, as in the throw would go up to 3 or 4 inches, and then come back down to 2.5 inches for several trials, and then go back up to 4 inches.

3. I noted that since my ball cleaner has been using Novus 1 cleaner for a long time, the padding may be impregnated with residual Novus, which could impact the results of the Aramith test (item 1 above). So I cleaned more balls by hand with Aramith and only buffed them until shiny with a cotton cloth. I did not put them in the ball machine. In this case, the throw was consistent at 5 inches the entire time (over 20 tries).


There is a clear difference in throw between using Aramith cleaned by hand and Novus 1 cleaned in a ball machine. The Aramith throws very consistently at 5 inches, even from the very first shot, whereas the Novus 1 throws inconsistently, and changes dramatically after fewer than 10 rolls up and down the table (10 trials). Novus "tops out" at a throw around 4 inches, but is still inconsistent well after 20 tries.

It seems to me that Aramith is the way to go since it provides consistency. I find that when using Novus 1, the balls open up more easily than when using Aramith. I wonder whether this is an "unfair" advantage, akin to spraying silicone on the balls. I would be interested to hear what is being used in tournament play. I believe it is wise to learn how to play under various playing conditions, but knowing how the balls were cleaned can help a player gauge how much the throw conditions are likely to change.


Is there any rationale behind the choices for pool balls?

Every "solid" (balls 1-7) have a different color, and every "stripe" (balls 9-15) uses the same color as the corresponding "solid" (e.g., the 1-ball and 9-ball are both yellow). If you add 8 to a solid number, you get the corresponding stripe number of the same color.

If you arrange the balls in an 8-ball rack in a methodical way in color groupings, all sorts of numerical consistencies also arise. For more info, see: "Ball Colors and Numbers" (BD, May, 2017).


from Patrick Johnson:

pool ball colors

Three primary colors:
1 & 9 = Yellow (primary)
2 & 10 = Blue (primary)
3 & 11 = Red (primary)

Three secondary colors:
4 & 12 = Purple (blue+red)
5 & 13 = Orange (red+yellow)
6 & 14 = Green (yellow+blue)

One tertiary color:
7 & 15 = Maroon (purple+red)

Two all-or-nothing colors:
8 = Black
Cue = White


compression effects

Does the size of the contact patch between the CB and OB vary with cut angle and speed, and does this affect how I should aim shots?

The CB and OB do compress a small amount during the extremely brief contact time, and this does create a "contact patch" that varies in shape and size with cut angle and speed, but the effect of this compression is very small. The compression does increase the cut angle slightly due to the contact patch compressing past the initial point of contact between the balls (when contact is first initiated), and this does tend to counteract the effect of cut-induced throw (CIT) a small amount, but the ball-compression effect is very small and can be neglected for all practical purposes.


contact time

How long are the CB and OB in contact during a collision?

Marlow did some experiments on this, and he reported numbers between 0.0001 and 0.0006 seconds (in "The Physics of Pocket Billiards," 1995). The balls do not stay in contact for very long.

Alexander Sorokin has also taken various measurements of ball properties and contact times. More info can be found here: Collision of Billiard Balls. They measured ball contact times in the 250-300 micro-second range (0.00025-0.0003 seconds).

Here's a super-slow-motion view showing how brief the balls remain in contact (and how little they compress):


cue ball types

What are the differences among commonly used cue balls?

from sfleinen (in AZB post):


from Michael Gaughan, Valley-Dynamo Parts Manager (via e-mail), in response to sfleinen's quote above concerning "Red Dot" balls:

In 1993 Dynamo began phasing out this ball in favor of a magnetic separator; and by 1995, all Dynamo coin tables used magnetic separation.


from DogsPlayingPool (in AZB post):

Aramith CB information

According to this the Red Logo, Blue Circle, and Measle are all the same ball. They are made from the same resin (Super Aramith Pro) and obviously are all regulation size and weight.


from sfleinen (in AZB post):

What's truly bizarre, is that only several years ago, all three of these cue balls were indeed different -- different weights and different compositions. You can just *look* at each of these three balls, and see the differences in opacity and color, nevermind the obvious differences you notice when you put the cue balls on the scale.

What I think happened of late, is that the Saluc factory, in the spirit of absorbing their "acquisitions" (e.g. Brunswick's and Aramith's products), have decided to standardize all the products.


from sfleinen (in AZB post):

... a lighter cue ball [is] undesirable in 14.1 (e.g. because of reduced ability to "hold" it [it "wants" to zing around], as well as the propensity to glance off the rack and clusters at weird angles).


ball elasticity

What is ball elasticity, and how does it affect how pool balls play?

Elasticity (described by the physics term "coefficient of restitution") is a measure of how much energy is retained during a collision between two balls. More information, including both technical details and illustrations on the affects on play can be found here:

ball condition effects resource page

from Bob Jewett (in AZB post):

material and manufacturing

What are pool balls made of?

Here are some good resources explaining the history and current state of pool ball materials:


Here's a basic description of the manufacturing method:


Some useful information is also available here:



Silicone spray effects

Does Silicone spray help you get more draw and massé action on the CB?

Yes. In fact, some trick shot artists sometimes use Silicone spray to help them create some of the magical shots they can execute.

I describe and demonstrate the effects of Silicone spray in this video (starting at the 0:40 point):

NV B.40 - Massé-draw billiard (carom) trick shot from the movie "The Hustler"

Here's another (at the 0:35 point):

NV B.41 - Coriolis massé shot aiming method with a large-curve example

It works like a charm, but it does wear off over time as it wipes off onto the cloth. It also leaves slippery residue on your hands when you handle the CB.


smoothness and roundness

What does a pool ball's surface look like under extreme magnification, and how smooth and round is a pool ball compared to the Earth?

Here's a 1mm x 1mm area of a clean and smooth cue ball imaged with a scanning white light interferometer:

magnified cue ball surface

The plentiful scratches are from the polishing process during manufacturing. The blemishes are simply slight defects in the formed material. The following contour map shows the relative size of the small peaks and valleys on the surface:

magnified cue ball surface contours

The scale indicated on the image above is microns (micrometers = 10^-6 meters = 0.001 mm).

See "Is a Pool Ball Smoother Than the Earth?" (BD, June, 2013) for more images and analysis. The quote from Bob Jewett below also offers a good interpretation of the data. And here's another good article on this topic: Is the Earth Like a Billiard Ball or Not? And if you think the Earth is flat, you need to watch this video: Is Earth Actually Flat?


from Bob Jewett (in AZB post):

So from Dr. Dave's picture we see a difference in heights of 1 micron (peak to valley) for spots that are within 1000 microns (1mm) of each other. Since the radius of a pool ball is about 28560 microns, the "local" roughness observed is about 1/30000 or about roughly 30 parts per million.

For a similar ratio on the surface of the Earth, consider the extreme of Mt. Whitney to Death Valley, which are pretty close to each other and differ in elevation by about three miles. Since the radius of the Earth is about 4000 miles, the "local" roughness of the Earth is about 1/1400 or 700 parts per million.

Measurements in Louisiana will give a different roughness. Driskill Mountain is 535 feet (or 0.1 mile, close enough) and the highest point in the state. 0.1/4000 is 1/40000 or slightly smoother than the polished ball Dr. Dave showed. If you look just at the oceans, the roughness is around 150 feet in the worst swells, but that would be quite a bit smoother than the polished pool ball.

Gouged pool balls -- from having been knocked onto the floor or against the wall -- might have pits 100 microns deep. Those would be rougher than California by a factor of 5 or so.

In my experience, Aramith balls out of the box are within 0.001 inch of the correct size and are rounder than that. That means that the non-roundness of a ball is less than 0.001/1.125 or 0.1% or about 1000PPM. The Earth is about 3300PPM out of round according to the above postings.

Bottom line: New, polished pool balls are much rounder than the Earth and somewhat smoother than the "geologically interesting" areas of the Earth. Old, worn pool balls are still much rounder than the Earth but depending on damage may be rougher than the roughest spots on the surface of the Earth.


from looey (in AZB post):

These are 3 photo's of a billiard ball's surface under microscope. I apologize for not having the actual magnification multiples (the mags were huge).

cue ball magnified

cue ball magnified

cue ball magnified


types for different games

What type and size balls are used in different cue sports games?

See ball types for different games.


weight, size, and wear effects

Does the weight and size of the balls ever vary much, and does it have an effect?

The following video describes and demonstrates all effects related to using cue balls or object balls of different weights and sizes:

"Ball Weight and Size Difference Effects – Part I" (BD, February, 2012) and "Ball Weight and Size Difference Effects – Part II" (BD, March, 2012) also cover ball-weight-difference effects in detail. Here's another article from Bob Jewett (BD, December '05) on the topic.

Generally, with older balls, the cue ball (CB) will be slightly smaller and lighter than the object balls (OBs) because it takes more abuse and wears faster as a result. However, if a new CB is used with an older set of OBs, the CB will be slightly heavier because only the OBs will have wear. On many coin-operated tables in bars (i.e., "bar boxes"), the CB is often heavier and/or larger than the other balls to help the ball-return mechanism distinguish the CB from the others.

When the CB is heavier, it is easier to follow and tougher to draw. With a cut shot, the CB will go forward of the tangent line; and with a stop shot, the CB will drift forward some. This effect is called "smash through." A heavier CB will also squirt slightly less.

When the CB is lighter, it is easier to draw the CB and tougher to follow. With a cut shot, the CB will pull back from the tangent line; also, with a stop shot, the CB will bounce back some. A lighter CB will also squirt slightly more.

When the CB is smaller or larger, the contact point on the OB will not be at the equator, the balls will also tend to hop a little, especially with faster speed (e.g., on an old bar box with a large CB, it is very easy to launch the CB off the table on a power break shot). With cut shots, the cut angle will also be off slightly (see the question and answer below), but this is an extremely small effect.

Worn OBs will also not rack as well as new high-quality balls. Slight mismatches in size and non-spherical shape (due to non-uniform wear) will result in less-tight racks and poor break action (bad spread, more clusters, fewer balls made).


Do CB and OB weight and size differences affect cut angle and throw?

See the throw ball-size effects resource page.

Assuming the ball surfaces have the same friction properties in a comparison, and assuming the same line-of-centers hits are being created in a comparison, then the amount of throw should not vary with CB or OB weight or size. However, if the weight difference is due to size differences, and a person aims the same way they normally do (with equal-size balls), then there will be different amounts of perceived throw. For example, a larger CB will create a slightly fuller hit than expected (with normal aim), and this will give the perception that the OB is being thrown more (even though it isn't).

With a larger CB, there is a downward component of force (which can make the CB and/or OB hop and result in slower OB motion), but this would not change the OB direction (i.e., the amount of throw). Interaction between the bottom of the OB and the cloth has nothing to do with throw. Although, having the CB contact the OB above the equator does change the effective cut angle of the shot, just as it does with jump shots where the CB hits the OB while airborne (and some people might perceive this as a throw effect, but it isn't). For more info and demonstrations of overcutting with an above-equator hit, see the jump-shot over-cut resource page.

Because the ball material is so stiff compared to the cloth, a downward CB-to-OB collision at a modest angle is really unaffected by the resulting OB-to-table collision. Per the ball-contact-time resource page, a ball-to-ball collision occurs in about 0.0003 seconds (300 microseconds). Per HSV B.44 - cloth compression and cue ball trajectory for draw shots of various elevations, an OB takes about 0.002 seconds to compress the cloth and rebound off the slate at an angle ... about 7-times longer than the ball-ball collision. Therefore, the CB-to-OB collision is mostly done before any significant force builds up between the OB and the cloth.

Even though a larger CB will hit the OB above the equator, which increases the effective cut angle some (creating a thinner hit), the larger size of the CB causes a sooner hit, which decreases the effective cut angle some (creating a fuller hit). The 2nd effect is bigger than the first, creating a fuller hit than expected. If you are not convinced, draw a top view of a half-ball hit with both an equal-size CB and a larger CB (along the same CTE line). The point of contact between the balls must lie on the line-of-centers between the balls. The line-of-centers for the larger ball creates a smaller cut angle and fuller hit (even with the above-equator hit effect). Although, any reasonable and typical size differences between the CB and OBs will probably be too small to notice any cut-angle-change effects, unless you are playing on an old "bar box" with a large CB, in which case it might be noticeable to a good player.


top of page      |     main FAQ index     |     website home page