ISETBIO Baseball Simulation Experiment - Revision history

Alipman1: /* Conclusions */

2025-12-12T23:29:41Z

Conclusions

← Older revision		Revision as of 23:29, 12 December 2025
Line 172:		Line 172:
	== Conclusions ==		== Conclusions ==

	This project ~~was really about getting~~ a first ~~look at~~ what the cone-level retinal signal might ~~contain during a~~ pitched ~~baseball,~~ under a controlled ~~setup~~. ~~Everything in this simulation happens~~ in ~~a vacuum given~~ the ~~data at disposal for this projects. So the results should be interpreted as a first pass at what~~ a ~~hitter~~ might ~~experience when facing live pitching, rather than taken verbatim. In this analysis,~~ a ~~few patterns emerged, but they should all be taken within that context~~.		This project should be viewed as a first-pass exploration of what the cone-level retinal signal might look like for pitched baseballs under a tightly controlled, highly simplified environment. The goal wasn’t to recreate real hitting conditions, but to get an initial sense of how seam orientation, velocity, and movement translate into differences in the retinal input a batter might receive during a pitch.

	~~Across all pitch types~~, seam orientation ~~changed the~~ foveal ~~cone response once the ball was roughly within~~ the last ~~5–10~~ feet. ~~Earlier in flight, the orientations were basically indistinguishable at~~ the ~~retinal level. Even where differences appeared late~~, they ~~varied~~ a ~~lot by pitch type. Changeups showed the strongest differences, sliders and curveballs sat~~ in the ~~middle, and fastballs barely shifted at all~~.		Even in this idealized setup, it was interesting to see how certain visual cues—especially seam orientation and pitch speed—modulate foveal activity in the last few feet of flight. While these simulations are far from capturing the real complexity of human vision during hitting, they offer a starting point for thinking about how early in flight the retina even has access to information that could help distinguish pitch types.

	~~The pitch-classification results mostly line up with intuition — it’s obviously easier to tell what pitch you’re getting when~~ the ~~ball is 15–20 feet away compared~~ to ~~50 feet out~~. ~~One~~ of the ~~main limitations of this analysis is that~~ the classifier was trained with each distance bucket being treated completely in isolation. A real hitter doesn’t work that way. When they see a pitch, they’re constantly updating their internal guess using everything they’ve already seen up to that point. A more realistic model would need to account for that accumulated information over ~~time, not just look at single snapshots. That’s something future versions should absolutely incorporate, however it is not something I was able to complete within~~ the ~~timeline~~ of ~~this project.~~		A more complete next iteration would build directly on this simulation pipeline: incorporating more realistic rendering, expanding the dataset to hundreds or thousands of trajectories, and introducing richer peripheral and temporal metrics. Those additions would help paint a fuller picture of how the entire retina—not just the fovea—responds over the course of a pitch, and how those signals evolve in a way that might matter to a real hitter.

	~~Taken together, the safest interpretation is that this simulation provides~~ a ~~baseline~~, ~~not~~ a ~~behavioral prediction. It shows what the cone mosaic~~ might ~~encode under idealized conditions, stripped of everything that makes real vision messy. It highlights where seam orientation could plausibly~~ matter ~~at the retinal level, but it does not tell us how~~ a hitter actually perceives or uses this information. Eye movements, neural processing, context, expectations, and higher-order visual pathways all play major roles that are completely absent here. However, I think this simulated study can be used as a baseline for future research to attempt to further research these effects.

	== Appendix ==		== Appendix ==

Alipman1: /* Seam Orientation foveal effects */

2025-12-11T20:12:58Z

Seam Orientation foveal effects

← Older revision		Revision as of 20:12, 11 December 2025
Line 136:		Line 136:
	[[File:foveal impacts.png\|center\|600px]]		[[File:foveal impacts.png\|center\|600px]]

	One question that came up during analysis was whether the stronger seam-orientation effects I saw in changeups were actually driven by differences in velocity or spin rate, rather than seam orientation itself. To isolate the role of velocity, I examined the signed β(v0) coefficients across all pitch types, seam angles, and distance bins. The pattern seems to follow that most β(v0) values are negative, indicating that higher velocities suppress foveal cone activity, while slower velocities enhance it. The effect is strongest inside ~15 ft of home plate, where slower pitches spend more time in the foveal window and produce higher photon catch, while faster pitches suffer additional motion-blur and reduced local contrast. Therefore the part of the “changeup advantage” in seam-related effects is simply likely that the batter’s fovea has more time to resolve the ball in that window of time compared to the faster pitch types.		One question that came up during analysis was whether the stronger seam-orientation effects I saw in changeups were actually driven by differences in velocity or spin rate, rather than seam orientation itself. To isolate the role of velocity, I examined the signed β(v0) coefficients across all pitch types, seam angles, and distance bins. In the graph below, blue suggests that slower speeds cause greater foveal activity, while red colors suggest faster speeds increase foveal activity. The pattern seems to follow that most β(v0) values are negative, indicating that higher velocities suppress foveal cone activity, while slower velocities enhance it. The effect is strongest inside ~15 ft of home plate, where slower pitches spend more time in the foveal window and produce higher photon catch, while faster pitches suffer additional motion-blur and reduced local contrast. Therefore the part of the “changeup advantage” in seam-related effects is simply likely that the batter’s fovea has more time to resolve the ball in that window of time compared to the faster pitch types.

Alipman1: /* Results */

2025-12-11T20:11:31Z

Results

← Older revision		Revision as of 20:11, 11 December 2025
Line 136:		Line 136:
	[[File:foveal impacts.png\|center\|600px]]		[[File:foveal impacts.png\|center\|600px]]

	One question that came up during analysis was whether the stronger seam-orientation effects in changeups were actually driven by differences in velocity or spin rate, rather than orientation itself. To ~~check this~~, I ~~fit models that included angle × spin~~, ~~angle × velocity~~, and ~~angle × spin × velocity interaction terms within each 5-ft~~ distance ~~bin~~. ~~As the heatmap below shows~~, ~~these interaction terms stayed near zero across all pitch types~~, ~~with no consistent brightening even~~ in the ~~late~~-~~flight region where~~ the ~~main~~ seam-~~orientation~~ effects ~~appear. This indicates~~ that the ~~retinal sensitivity~~ to ~~seam orientation is essentially independent~~ of ~~spin rate and velocity in this simulation, instead,~~ the ~~effect arises from geometric visibility (i.e., how the seams face the viewer) rather than from dynamic factors like RPM or initial speed~~.		One question that came up during analysis was whether the stronger seam-orientation effects I saw in changeups were actually driven by differences in velocity or spin rate, rather than seam orientation itself. To isolate the role of velocity, I examined the signed β(v0) coefficients across all pitch types, seam angles, and distance bins. The pattern seems to follow that most β(v0) values are negative, indicating that higher velocities suppress foveal cone activity, while slower velocities enhance it. The effect is strongest inside ~15 ft of home plate, where slower pitches spend more time in the foveal window and produce higher photon catch, while faster pitches suffer additional motion-blur and reduced local contrast. Therefore the part of the “changeup advantage” in seam-related effects is simply likely that the batter’s fovea has more time to resolve the ball in that window of time compared to the faster pitch types.


	[[File:~~seamo terms~~.png\|center\|600px]]		[[File:velo effect.png\|center\|600px]]

	=== Pitch Discrimination ===		=== Pitch Discrimination ===

Alipman1: /* Results */

2025-12-11T19:34:39Z

Results

← Older revision		Revision as of 19:34, 11 December 2025
Line 139:		Line 139:


	[[File:~~new~~ terms.png\|center\|600px]]		[[File:seamo terms.png\|center\|600px]]

	=== Pitch Discrimination ===		=== Pitch Discrimination ===

Alipman1: /* Seam Orientation foveal effects */

2025-12-11T19:32:15Z

Seam Orientation foveal effects

← Older revision		Revision as of 19:32, 11 December 2025
Line 139:		Line 139:


	[[File:~~seamo~~ terms.png\|center\|600px]]		[[File:new terms.png\|center\|600px]]

	=== Pitch Discrimination ===		=== Pitch Discrimination ===

Alipman1: /* Seam Orientation foveal effects */

2025-12-11T19:30:56Z

Seam Orientation foveal effects

← Older revision		Revision as of 19:30, 11 December 2025
Line 140:		Line 140:

	[[File:seamo terms.png\|center\|600px]]		[[File:seamo terms.png\|center\|600px]]


	=== Pitch Discrimination ===		=== Pitch Discrimination ===

Alipman1: /* Pitch Discrimination */

2025-12-11T18:52:06Z

Pitch Discrimination

← Older revision		Revision as of 18:52, 11 December 2025
Line 169:		Line 169:
	[[File:correlation matrix1.png\|center\|400px]]		[[File:correlation matrix1.png\|center\|400px]]

	The ~~plot above~~ shows ~~how~~ four physical pitch ~~characteristics: spin rpm~~, ~~ball~~ speed (v0), horizontal break (hb), and vertical break (hb) ~~correlate to~~ the probability of the ~~eye discriminating~~ the ~~pitch type~~. The ~~thing that stands out the most to me,~~ is the ~~relation~~ between spin rate and ~~velocity on pitch type~~ classification~~. Across~~ changeups, ~~fastball~~, and sliders, ~~as spin rate increases, so does the probability that the pitch will be classified correctly~~. Because ~~both~~ spin ~~rate~~ and ~~velocity impact~~ movement (~~horizontal~~ and ~~vertical break~~), ~~this tells me~~ that ~~this greater separation of~~ movement ~~ultimately allows~~ the ~~batter to discriminate pitches more effectively~~.		The table shows, for each pitch type, the correlation between four physical pitch characteristics—spin rate, initial speed (v0), horizontal break (hb), and vertical break (vb)—and the Random Forest’s probability of assigning the correct pitch label based on the retinal features. Red cells indicate traits where higher values are associated with greater classifier confidence (higher true-class probability), and blue cells indicate traits where higher values are associated with lower confidence. The clearest pattern is the positive association between spin rate and classification confidence for changeups, fastballs, and sliders (r ≈ 0.26–0.49), with a similar positive effect of velocity for fastballs (and to a lesser extent changeups). Because spin and speed jointly shape the amount and direction of movement (hb and vb), a reasonable interpretation is that pitch instances with more distinctive movement profiles generate more separable retinal signals, which in turn makes them easier for the model—and by extension, potentially the visual system—to tell apart.

	== Conclusions ==		== Conclusions ==

Alipman1: /* Pitch Discrimination */

2025-12-11T03:13:13Z

Pitch Discrimination

← Older revision		Revision as of 03:13, 11 December 2025
Line 167:		Line 167:
	Another interesting way to look at these results is to examine how physical characteristics relate to pitch discrimination using retina characteristics.		Another interesting way to look at these results is to examine how physical characteristics relate to pitch discrimination using retina characteristics.

	[[File:correlation ~~matrix~~.png\|center\|400px]]		[[File:correlation matrix1.png\|center\|400px]]

	The plot above shows how four physical pitch characteristics: spin rpm, ball speed (v0), horizontal break (hb), and vertical break (hb) correlate to the probability of the eye discriminating the pitch type. The thing that stands out the most to me, is the relation between spin rate and velocity on pitch type classification. Across changeups, fastball, and sliders, as spin rate increases, so does the probability that the pitch will be classified correctly. Because both spin rate and velocity impact movement (horizontal and vertical break), this tells me that this greater separation of movement ultimately allows the batter to discriminate pitches more effectively.		The plot above shows how four physical pitch characteristics: spin rpm, ball speed (v0), horizontal break (hb), and vertical break (hb) correlate to the probability of the eye discriminating the pitch type. The thing that stands out the most to me, is the relation between spin rate and velocity on pitch type classification. Across changeups, fastball, and sliders, as spin rate increases, so does the probability that the pitch will be classified correctly. Because both spin rate and velocity impact movement (horizontal and vertical break), this tells me that this greater separation of movement ultimately allows the batter to discriminate pitches more effectively.

Alipman1: /* References */

2025-12-10T04:56:47Z

References

← Older revision		Revision as of 04:56, 10 December 2025
Line 196:		Line 196:

	‌[5] Grigorios Chrysos, et al. “Motion Deblurring of Faces.” International Journal of Computer Vision, vol. 127, no. 6-7, 1 June 2019, pp. 801–823, https://doi.org/10.1007/s11263-018-1138-7. Accessed 22 Apr. 2023.		‌[5] Grigorios Chrysos, et al. “Motion Deblurring of Faces.” International Journal of Computer Vision, vol. 127, no. 6-7, 1 June 2019, pp. 801–823, https://doi.org/10.1007/s11263-018-1138-7. Accessed 22 Apr. 2023.

			[6] Kagan, David, and Alan M. Nathan. “Statcast and the Baseball Trajectory Calculator.” Physics Teacher, vol. 55, no. 3, 2017, eric.ed.gov/?id=EJ1142622, https://doi.org/10.1119/1.4976652.

Alipman1: /* Appendix */

2025-12-08T07:10:47Z

Appendix

← Older revision		Revision as of 07:10, 8 December 2025
Line 182:		Line 182:

	== Appendix ==		== Appendix ==
			Code: https://github.com/alexlipman13/pitch-trajectory-optics

	== References ==		== References ==