4/26/2016 Updates
Interface Improvements
Faster Rendering
People have been making larger and larger models. We've made this a lot easier by making several advancements to editor speed. Simulations are now performed in web workers, and Guesstimate now waits for you to pause while typing before propagating changes through a model.
Bold Outputs
When looking at new models, it's important to get a quick idea of what matters. We've made all output nodes bold, and intermediate nodes and disconnected nodes are slightly darkened.
The 'to' Syntax
We've witnessed a fair bit of confusion over the previous ± syntax. Often the bounds on the right and left sides are very different, so averaging them led to surprises, especially for skewed distributions. We've replaced this with a low to high range, which displays a 90% confidence interval that always contains the mean.
Clear Selection Highlighting
We've heard reports that it wasn't obvious which metric was selected. To help, we've changed the selection highlighting to be more clear. Note that when a metric is selected (but not a text area within the metric), you can press backspace to delete it.
New Features
Discrete Distributions
Guesstimate now supports the Poisson, Bernoulli, Binomial, and Negative Binomial distributions.
Financial Functions
Interested in estimating Compound Interest or IRR? Now we have the functions to support them. We support the following functions, plus several other financial functions. Read about them in the Functions: Finance section of our sidebar documentation.
- Amortization
- Compound Annual Growth Rates
- Compound Interest
- Discount Factor
- Future Value
- Net Preset Value
- Weighted Average Cost of Capital
Guesstimate Scratchpad
Making a quick calculation that you don't want to save? You may be interested in using the scratchpad (opens in a new tab), a tool accessible to everyone (including logged out users). This comes with some instructions in how to use Guesstimate. You don't even have to be logged in to use it.
Team Accounts Improvements
We still have team accounts in private beta, but have been expanding their functionality. If you're interested in collaborating on models with your team, please let us know in an email.