Watch Data

Data Completeness

Barker's model used for valuation is a supervised machine learning model and therefore is highly dependent on the quality and comprehensiveness of the input data. For example, every watch at auction we collect information on the watch, the auction, valuation estimates, final sale prices, and more.

For watches, Barker has built an understanding of the market by collecting a comprehensive dataset consisting of 800 thousand watch sales from 2012 onwards.

Influences on Watch Valuation

Barker has found the features which provide the greatest uplift in predicting the price of a watch. The features used in the machine learning model include:

Watch Identification:

• Make

• Reference No.

• Family

Watch Characteristics:

• Condition

• Year Made

• Materials

• Tropical-Dials

Watch Accessories:

• Band type

• Certification papers

• Original Box