I had a fellow woodworker leave a comment about rake and splay on my last post My Use of SketchUp as an Aid and I wanted to address it and see if I can help them out. This will also provide me (and anyone else who wants it) some teaching material for my students when we discuss it. I usually use the whiteboard and draw and talk while showing physical examples but this collection of pictures will allow me to save it as a Powerpoint presentation.
How important is rake and splay? The real answer is: sort of. There are hundreds, if not thousands, of chairs out in the wild that lack either rake or splay or both and they’re doing fine. This post will hopefully enlighten you on some concepts to think about while designing and building your chair or stools or teaching students to build.
The science behind rake and splay: the primary force that we’re concerned about is the downward force of the body weight of the sitter. We size, position, and angle the legs accordingly so when the person is sitting, the legs will hold them up. This is a pretty straightforward strategy.
Rake is the forwards and backwards angle as the person sits. Splay is the left and right angle as the person sits.
One method that I use to help illustrate different principles that I teach, is to use the extremes. Even though it may seem ridiculous, it helps us to visualize the concepts and understand limitations.
We have three examples of rake and splay. Chair A. is a design with moderate rake and splay. Chair B. is a design with no rake or splay. Chair C. is a design with extreme rake and splay. Here are the front and side views.
Without getting into it, you probably can already identify issues and design concerns. I'm a firm believer that people have more knowledge than they realize about furniture but often don’t know that because they aren’t actively thinking about it. When they’re presented with extremes or poor designs, they can usually identify it, even if it’s only to say that “it doesn’t look good/correct”.
If we were to use a red arrow to represent the downward force of a sitter, we can imagine that a chair with no rake and splay would still be okay to use if the force is applied within the footprint of the legs of the chair. This is easy to accomplish as the center of our mass should be aligned with the center of the chair.
The concern arises when the force changes from a strictly downward, perpendicular force and starts to become a diagonal or sideways force as the sitter leans, especially if it’s on the outer edge of the seat AND past the footprint of the feet of the leg as shown below.

As we can see with this example, the splay (left to right leg angle) keeps the force being applied to the edge of the chair within the footprint of feet of the legs.
Not all chairmakers are concerned with splay as seen with many post and rung (ladderback) chairs. Jimmy Possum chairs, a chairmaker from Australia also lack splay. The way that these chairs are constructed, giving the legs splay would make the arms too narrow for the sitter. Fascinating construction and idea though.
Hey Josiah, what about someone leaning back in their chair?
This is where we see the benefits of rake. In the following picture, we can see that as long as the majority of the force is applied at point A or forward, it’s still within the footprint of the legs of the chair due to the legs being raked moderately backwards. It works out for us that even with the sitter leaning back, the majority of the downward force is applied to the chair by their m(ass) on the seat.
If the force were to suddenly and drastically change to point B, we would be concerned and the foot of the back leg would become a fulcrum and the chair could tip. Where might we see this sudden and drastic change? We see this with little kids or drunk adults, standing in the chair and applying force to the back of the chair. In this chair example, to make the chair “safer” our options would be to increase the rake of the back legs even more, decrease the angle or the back sticks a little, or monitor our children and drunk friends more closely.
When designing the rake of the front legs, we’re not as concerned about a sitter tipping the chair forward so the rake of the front legs would be focused on the tripping hazard of someone passing in front of the chair or the sitter maneuvering into the chair. They tend to be rather conservative, aligned with or just pass the front edge of the chair by a couple/few inches.
How much is too much rake and splay? Chris Schwarz thinks that we tend to be more conservative than needed. Going back to our pictures, we can see that chair C has an extreme amount of rake and splay and we would worry about the integrity of the seat and leg tenons holding up to the force of the sitter. We wouldn’t want our sitter to have the same experience that Mel Gibson had with a chair that he made in the opening scene of The Patriot.
What is the magical number for too much rake and splay? I’m not sure, I haven’t reached them yet and we don’t have any historical examples because we don’t tend to keep failures to have the measurements. I would use historical examples and other chairmaker’s work to help me to decide but hey, sometimes we need to experiment on our own.
Here’s an example of one I found in an article that Chris Schwarz wrote about American Stick Chairs with extreme rake and splay. It’s an early American Windsor style comb-back chair.
Now, what about leg placement, where should they go? In this picture we see the example on the right goes to the extreme, pulling the legs towards the center of the seat. When we design the leg placement, we don’t want the sitter to feel like they’re in a circus act, trying to balance themselves. We also don’t want to put them dangerously close to an edge or in the way of the short/long sticks.
Besides understanding that, there aren’t any hard numbers dictating where they should be placed. As a beginner, follow and copy chairmakers that you respect, look at pictures on the interwebz, and start playing around with CAD software or sketches.
Which takes us from the science of rake, splay, and leg placement to the design of it found in Part 2.