A hypothesis is a stated proposition or assumption about a population, process, or relationship that can be evaluated with evidence. It gives an analysis a concrete question to investigate before scripts and tests begin. section 1.3
Use a hypothesis to guide an evidence-based workflow:
- State a null hypothesis or practical assumption. section 1.3
- Prepare, reshape, group, summarize, and visualize relevant data. section 6.0
- Apply an appropriate method, then reject, fail to reject, confirm, or deny the proposition according to the evidence. section 1.3
NBA draft position and career games
A hypothesis proposes that players selected earlier in the first round will play more regular-season games than later selections. Grouping draft picks into segments and comparing means, medians, and shares of total games supports the claim that teams moving up the draft board are more likely to select players who play in more games. section 3.5.1
Fourth-quarter wins
Another hypothesis challenges the belief that NBA games are won in the fourth quarter. Almost every regular-season game across two contiguous seasons is examined by comparing winning teams' fourth-quarter performance with their performance in the other three quarters. section 6.0
Figure 1.5 A typical chapter flow and, not coincidentally, the typical end-to-end flow of most real-world data science and statistics projects
Follow the end-to-end project flow from hypothesis through scripting, analysis, visualization, and interpretation. section 1.3
