Statistical modelling of training and performance using power output and heart rate data collected in the field

Al-Otaibi, NM

Abstract

This thesis develops statistical models of performance and training that make use of
power output and heart rate data. These data were collected during training and
competition, and were recorded every five seconds using a power meter and heart rate
monitor. Using these data, we estimate the parameters of the Banister model of training
and performance. In principle, knowledge of these parameters allows one to provide
quantitative decision support for the scheduling of training in advance of a major
competition.
The methodology proceeds in a number of steps. In the first, measures of both training
and performance must be specified. The training experienced by an athlete in a single
session, the training load, can be measured in a number of ways. We use the TRIMP
measure. This measure in its simplest form is essentially the total number of heart beats in
a training session. Then the training loads of successive sessions are accumulated into a
single measure of training up to time t. This we term the accumulated training effect (at
time t). Performance during a session at time t is defined as a function of the power output
observed during the session. We consider various performance measures and describe
these in detail in the thesis. Then in the second step, we relate the performance at time t to
the training load up to time t using a regression model, estimating the parameters of the
performance training relationship. The final step is the training optimisation step, whereby
the known training-performance model parameters can be used to specify training loads up
to time T that will maximise (in expectation) the performance at time T.
We demonstrate the methodology using the training data histories of ten competitive
male cyclists. As each athlete has his own specific characteristics, we should focus on
optimising training and performance individually. We compare and contrast the different
performance measures that we propose.
Our principal findings are that: Banister model parameters can be estimated; that the
different performance measures yield different Banister model parameter estimates and
therefore that the performance measure specification is a matter for athlete/coach choice;
and that finally the Banister model has a serious shortcoming for the optimisation of
training. The articulation of this shortcoming is an important contribution of this thesis

Citation

Al-Otaibi, N. (in press). Statistical modelling of training and performance using power output and heart rate data collected in the field. (Thesis). University of Salford

• Thesis Type: Thesis
• Acceptance Date: Feb 1, 2017
• Deposit Date: Sep 19, 2018
• Publicly Available Date: Oct 19, 2018