Physics and Rowing

1) -"There is not much Physics in Rowing"

1a The Physics of Rowing

Chris Pulman
Gonville & Caius College, University of Cambridge

Relatively little is known about the exact mechanisms that explain the motion of rowing boats.
This essay discusses some of the physical processes that govern the motion of eight-oared
boats (VIII). Boats move because momentum is transferred to the water by the rowers and their
oars, causing the boat to move forwards; a process complicated by the motion of the rowers
within the boat itself. Many forms of drag are manifest in resisting the motion of the boat such
as skin, form, wave, aero-dynamical and inertial. An analysis of each concludes that it is
usually inertial drag which has the greatest effect, although both wave and aero-dynamical
drag can also be significant, depending on the weather conditions and speed of the boat.
Moreover, due to the resistive drag, in order to double the boat speed eight times as much
effort is required from the rowers. Although classical ship analysis shows that the VIII will be
statically unstable, an appreciation of the flow of water around the bows provides an
explanation for the stability of moving boats. Finally, the mechanisms of the rowing stroke are
investigated. It appears that lift forces – acting in a horizontal direction - provide the main
means of acceleration in the early and late stages of the stroke, while it is the movement of
water by the blade which imparts momentum to the boat during the middle part of the stroke.

I. Introduction II. Propulsion

This essay presents a discussion of the
physical processes that govern the motion of
eight-oared boats (VIII), although the physics
may also apply to other rowing boats. An VIII
is a narrow boat (70cm maximum width, 20m
in length) powered by eight rowers each
holding an oar, steered by a coxswain. The
rowers are mounted on sliding seats with their
feet attached to the boat by restraints.
The basic principle of rowing is quite
simple; momentum is transferred to the water
by pulling on the oar and pushing with the
legs, which causes the seat to slide backwards.
The oars pivot on “riggers” which lever the
water backwards. However, the motion of the
boat is complicated by the movement of the
rowers within the boat, resulting in it moving
fastest when the oars are out of the water and
the rowers are moving towards the stern of the
boat.
Opposing the motion of the boat is
drag which appears in several forms: skin,
inertial, aerodynamic and wave. The effects
and applicability of each type of drag are
considered.
The balance properties of an VIII are
discussed in the context of classical ship
theory and an appreciation of the flow of water
around the bows of the boat. Finally, the
rowing stroke is examined and the mechanisms
of lift and reversal of the oar blade, caused by
the movement of water, are compared.
Momentum is imparted to the water by the oar,
resulting in the boat accelerating in the
opposite direction. Although the blade appears
to “lock on” in the water, it must move the
water in the opposite direction to the motion of
the boat in order to conserve momentum. As it
is energetically efficient to move a large
amount of water slowly, rather than a small
amount quickly, blades have a large surface
area in order to maximise the water displaced.
The basic principles of the propulsion
of a boat are, however, complicated by the
variability of the centre of mass (CM) of the
rowers caused by the sliding seats (fig. 1).
When the oars are in the water, this relates to
the momentum imparted to the water. In the
recovery phase, when the oars are not in the
water and the rowers move towards the stern
of the boat, momentum conservation (equation
1) requires that the boat surges forwards. This
motion is clearly visible when viewing a
rowing boat.
FIG.1. (a) The crew is at “backstops”, the end of the
stroke, and the blade has just been removed from the
water; the crew and boat are at rest with respect to
each other so they move with the velocity of the CM,
vt. (b) The crew is sliding forwards to the stern of the
boat at a speed vc which causes the boat to surge
forward with extra velocity vb.
 m c v t + mb v t = m c (v t − v c ) + mb (v t + v b )
mc
⇒ vb = vc (1)
mb
mc, vc, mb and vb are the masses and velocities
of the crew and the boat respectively and vt is
the boat speed immediately after the blades
have been removed from the water.
The equations of motion of a rowing
boat (Brearley and de Mestre 1996) are, for the
power phase:
(mb + mc ) du = E (t ) − mc H1 (t ) − R (t ) (2)
dt
and the recovery phase:
(mb + mc ) du = mc H 2 (t ) − R(t )
dt
where u is the boat speed, E(t) is the effort
force exerted on the oars by the rowers, R(t) is
the total drag resistance to motion and H1(t)
and H2(t) are the accelerations of the rowers’
bodies during the power and recovery phases
respectively. Brearley and de Mestre suggest
that the motion of the rowers can be taken as a
half-cycle of simple harmonic motion.
Lazauskas (1997) showed that there are three
primary forms of E(t) (fig. 2) and computer
simulations (van Holst 2004) have shown that
the symmetrical form (fig. 2ii) is the most
energetically efficient. The drag resistance is
discussed in section III.
Effort E(t) is transmitted from the
rower to the water through the oars which, in
the frame of the boat, are class 1 levers (fig. 3).
The load is:
⎛b⎞
L = E⎜ ⎟
⎝a⎠
FIG. 2. Force curves E(t) for the power phase of the
stroke. (i) Peak pressure applied at the beginning of
the stroke; (ii) Peak pressure applied during the
middle of the stroke, with the effort symmetrical
about that point; (iii) Peak force applied at the end of
the stroke.
FIG.3. A class 1 lever, effort E is applied at the oar
handle and is transmitted through the fulcrum to the
load L at the blade.
The work done by the rower is transferred to
the water as kinetic energy, causing it to move
past the boat. However, the observer on the
river bank views the oar as a class 2 lever, with
the blade acting as a fulcrum and the boat
being levered past the stationary blade (fig. 4).
(3)
FIG. 4. A Class 2 lever. The effort E is applied at the
oar handle, but the blade acts as the fulcrum and the
load at the “rigger” (L) is moved.
III. Resistance
Like any body moving in a fluid, the motion of
rowing boats is opposed by resistive drag;
momentum is transferred to the water by the
passing of the boat. Characteristic types of
drag are skin, inertial, aerodynamic and wave.
Skin drag arises from the formation of a shear
boundary layer around the hull arising from
the viscosity of the water. This causes the
boat’s effective mass to increase as water is
dragged along by the boat. Inertial drag arises
due to the water in front of the boat being
accelerated, resulting in the boat’s momentum
decreasing. Wave drag represents the
dissipation of energy in the creation of waves.
The shape of the hull is a significant factor
(4)
since the drag is a function of the wetted area
of the boat.
FIG. 5. The top slab represents the boat moving at
velocity v at height h above the river bed. A shear
layer forms between the two which gives rise to a
velocity field caused by the water being dragged
along by the passage of the boat.
For boats moving in shallow water,
viscous drag is predominant (fig. 5). The fluid
adjacent to the boat forms a boundary layer
and a shear is set up in the fluid with a velocity
gradient v/h, modelling the bottom of the boat
as a flat sheet. The presence of a boundary
layer means that the boat is dragging water
along, increasing its effective mass. The
resistance per unit area acting on the boat due
to the fluid is then
R dv v for the drag resistance. There are also
=η =η
A dz h
However, this analysis applies only to shallow
water (<10cm).
v ≡ v1 ( z, t )
"∇"
FIG.6. Slab representing the underside of the boat.
The downward pointing arrow represents the direction
of the velocity gradient (ie "∇" ) while v1 represents
the velocity field of the fluid.
Generally, as the boat moves through
stationary water, the fluid in contact with the
bows is immediately accelerated to the boat
speed, v, but the shear layer only extends
downwards as the vorticity Ω diffuses away
from the boat (equation 6, fig. 6).
Ω = ∇× v
So the water is accelerated in accordance with
the diffusion equation (7).
∂v1 ( z , t ) ∂ 2 v1 ( z , t )
ρ =η
∂t ∂z 2
An approximate solution of this equation gives
the width ∆ of the boundary layer at a time t as
ηt
∆=
ρ
This indicates that resistance is proportional to
boat speed; however, within water, equation 8
implies that during one stroke the boundary
layer only diffuses ~2mm which is not
particularly significant.
In contrast, the high Reynolds number
of water implies that the primary resistance
impeding the motion of the boat is the inertial
drag. The pressure exerted by the fluid on the
1 2
boat is ρv where v is the speed of the
2
water relative to the boat (i.e. the boat speed).
This gives the drag resistance as
1
R= C D ρv 2 A
2
where A is the area of the boat exposed to the
stream flow and CD is the drag coefficient of
the boat, defined as the ratio of actual drag to
ideal drag. This gives more acceptable values
(5) implications for power requirements; the
momentum change of the water as a result of
the passage of the boat in a time t is:
∆p = ρAvt (9)
So the power required to maintain constant
velocity v is then:
t v
∫ Fdt = ∫
0 0
ρAvtdv
1
⇒F= ρAv 2
2
1
∴ P = Fv = ρAv 3 (10)
2
Thus, eight times more power is required to
double the boat speed. Indeed, this is actually
sensed when rowing.
In strong headwinds aerodynamic
(6) drag can severely resist the motion of the boat,
although its usual impact is small. Drag due to
air resistance has the same form as equation
10; however, the smaller density of the air
lessens its contribution. The velocity of the
boat is then replaced by sum of the boat and
(7) wind velocities. It is clear that for strong
headwinds air resistance rapidly becomes
significant. The drag is proportional to the area
exposed to the flow – i.e. the projected area of
the rowers into the wind, although this is a
poor approximation as each rower shields
those rowers behind, causing complex air-
(8) flows which affect the calculation.
FIG. 7. The length of the boat L is such that the waves
to the left of the boat are at a minimum. In contrast,
those at the right of the boat are at a maximum
resulting in resonant losses due to the pressure
difference between the two ends of the boat.
A rowing boat is narrow and is not
able to generate large waves. However, if the
length of the boat, L, is such that
2n + 1
L= λ (11)
2
(9)
where λ is the wavelength of the waves
generated, then resonant losses occur (fig. 7).
To avoid this situation the boat should not
travel at speeds
Lg
v= (12)
(2n + 1)π
At these speeds the energy dissipation due to
waves is maximal and so wave drag is
significant. The Froude number, F, indicates
the significance of wave drag and for an VIII
travelling at 5ms-1 (i.e. 2 km racing speed) it is:
v a wide-bottomed boat with a shallower hull-
F= = 0.35 (13)
gL
which corresponds to a local minimum in the
wave drag (fig. 8), so although wave drag is
certainly present, it is not significant for a
racing boat.
FIG. 8. Variation of the wave drag CD with the Froude
number F. A boat moving at race pace minimises the
wave drag locally.
IV. Balance of rowing boats
Racing boats have their centre of gravity (CG)
some distance above their centre of buoyancy
(CB – the CG of the submerged portion of the
boat). Since the boat is partially submerged, an
upthrust acts vertically through the CB and the
meta-centre (MC – the geometrical centre of
the boat). If the boat starts to roll, then a
couple is generated due to the weight of the
boat and rowers, which acts at the CG, and the
upthrust, which acts at the CB. If the CG is
below the MC, then the couple is in the
opposite direction to the roll, and the boat
returns to an upright position. If, however, the
CG is sited above the MC, the couple acts in
the same direction as the initial roll, and
provides positive feedback: the boat tips over
(fig. 9). The limit of stability is
I The function of the stroke is to transfer the
z= (14)
V the boat. The mechanisms behind this transfer
FIG. 9. When the CG is cited at the MC the balance is
neutral and the boat is stable. Similarly, when the CG
is sited below the MC the boat is also stable.
However, when it is sited above the MC a turning
moment is generated for any small displacement from
equilibrium and the boat will roll over (it is unstable).
Where z is the limiting distance of the CG over
the CB, I is the moment of inertia of the
waterline shape of the boat about the axis of
roll, and V is the volume of fluid displaced.
The CG for an VIII is approximately 10cm
above the seats, so equation 14 illustrates that
curvature (and larger moment of inertia), is
more stable than a narrow one, taking longer to
roll over. However, the increased surface area
produces a higher drag, so most racing shells
tend to be quite narrow. Although a stationary
empty boat has been shown to be stable,
measurements (Kerr 1996) show that for an
VIII with crew on board the CG is ~15cm
above the MC (just above the height of the
seats) causing inherent instability.
FIG. 10. (a) The boat is stable since there is a
balanced water flow exerting reaction forces on the
hull. (b) The boat is tilted and the unbalanced water
flow causes the boat to return to its level position.
Whereas instability of a static boat is
often observed; a moving boat tends to balance
well with little correctional input from the
rowers, such as leaning or altering the hand-
heights of the oars. As the bows move through
the water, around 400kg of water is displaced
per second for a boat travelling at 5ms-1.
Although much of this water is moved away
by the waves generated, significant forces
normal to the hull result due to the momentum
change caused by the passage of the bows. In
particular, the V-shaped bottom of the bows
produces turning moments which cause the
boat to return to the upright position (i.e. – the
moving boat is in a stable equilibrium, fig. 10).
The greater the volume of water displaced per
second – i.e. the greater the speed – the more
significant this effect becomes.
V. The rowing stroke
work done by the rowers into kinetic energy of
of momentum, however, are relatively
complex. Indeed, it has been a subject of
controversy within the rowing community as β
to the exact interaction of the blade with the
water. It is generally accepted that, at the
beginning and end of the stroke, lift – acting in
a horizontal direction - provides the main α
impetus for propulsion. However, during the
middle of the stroke the resistance of water on FIG. 12. The angle of attack β has been increased so
the blade provides propulsion. It is not clear that the longitudinal component propelling the boat is
which of these mechanisms is predominant. larger at the beginning of the stroke.

CD
= tan α (16)
CL

α As α increases throughout the stroke FD

increases and the lift decreases until α≈40o
when the blade stalls in the water – in the same
FIG. 11. Diagram of the oar. AB is the blade, u is the way an aeroplane stalls in the air - and lift is
velocity of the water, α is the angle between the oar zero. At this point resistance against the blade
and the side of the boat, vb is the velocity of the boat
and FR is the force acting on the blade resulting from through the exchange of momentum with the
the drag (FD) and the lift (FL) forces. water provides the mechanism for propulsion.
When α increases to approximately 110o the
At shallow angles α (fig. 11), the
blade again begins to experience lift until the
blade acts like a hydrofoil and the resultant
end of the stroke.
force on the blade is composed of a drag force
Young (1997) studied this sequence
FD and a lift force FL. As water is directed to
of events in detail. As the blade catches at the
one side of the blade a reaction force results -
beginning of the stroke a slow-moving vortex
the lift. In contrast to an aeroplane wing, the
is shed at the end of the blade. This results in a
flow of the water around the blade results in
bound vortex (with the opposite sense)
the lift force acting horizontally, rather than
forming around the blade, providing lift until
vertically. The lift does no work, so if it is
the blade stalls in the flow. At this point the
maximised while the drag force is minimised
fast-moving bound vortex around the blade is
then little energy is wasted and the boat moves
shed. At the end of the middle phase of the
more quickly. Under these conditionss,
stroke, a further fast-moving vortex is shed
however, the longitudinal component
from the end of the blade as another bound
propelling the boat is rather small. It has been
vortex (with the opposite sense) forms around
suggested by Brearley (1998) and van Holst
the blade providing lift. At the end of the
(2004) that altering the angle of attack (fig. 12)
stroke this is then shed (fig. 13).
would maximise this effect in the first part of
Interestingly, the blade tip moves
the stroke but have a negative effect in the
approximately 10cm in the direction of motion
latter part. If the rowers aim to produce peak
of the boat by the end of the stroke.
power at the beginning of the stroke altering
Furthermore, during the stroke it can reach as
this angle may provide an advantage, although
far forward as 40cm. Also, the observer on the
this is not energetically efficient (fig. 2). The
bank views the blade as being stationary when
drag and lift forces are:
the blade stalls. The blade velocity throughout
the stroke was shown to vary significantly;
1 when experiencing lift it was directed in the
FD = C D ρAv 2 (15)
2 direction of motion of the boat (with a
maximum velocity of 3ms-1). In contrast, when
1 the blade was stalled it moved backwards
FL = C L ρAv 2 (16) (with a maximum velocity of -1ms-1) which is
2 consistent with momentum exchange with the
water. Significantly, this implies that the
Where CD and CL are the drag and lift maximum dynamic pressure generated in the
coefficients and v is the velocity of the lift phase is nine times larger than that during
undisturbed flow. The directions of FL and FD the momentum exchange phase. However, the
are determined by the direction of circulation longitudinal component of this force is small
of the vortices. It can also be seen from fig. 11 during the lift phase but is large during the
that the CD and CL are related by: momentum exchange phase. Consequently, it
 α
α

α
α
FIG. 13. (a) At the beginning of the stroke a small leading-edge vortex is shed causing the creation of a lift-inducing vortex
(of the opposite sense) around the blade. (b) The angle α increases to the point where the blade stalls in the water so the
large vortex encapsulating the blade is shed. (c) The blade angle once again allows lift and so a large leading-edge vortex is
shed, forming a further counter-sense vortex around the blade. (d) At the end of the stroke the blade ceases to move and the
small vortex around the blade is shed. The result at the end of the stroke is two small slow-moving vortices formed at the
beginning and end of the stroke and a fast-moving twin-vortex system formed during the middle of the stroke. The
longitudinal velocity of the blade v relative to the water is shown in each figure.

is unclear which mechanism provides the
dominant contribution to propulsion.
It is clear, however, that lift is a more
efficient mode of propulsion since it involves
the movement of a large amount of water with
a small velocity, while propulsion derived
from resistance involves the movement of a
small amount of water at a large velocity. This
interpretation of the stroke helps to explain
why it is important for rowers to have a “fast
catch”; the faster the blade enters the water the
more quickly the leading-edge vortex can be
shed and lift provided to the blade. This also
applies at the “finish” where removing the
blade quickly prevents significant drag upon it.
The blade has stopped moving forwards
through the water so it has shed its vortex and
is susceptible to drag resistance; quick removal
minimises this effect.
VI. Conclusion
Although the motion of rowing boats appears
quite simple, it is, in fact, somewhat
complicated. Not only does the motion of the
rowers affect the speed of the boat, but also the
river conditions and the way the stroke is
taken.
It is suggested that for shallow rivers
skin drag and viscous effects will have a
significant contribution. However, in most
cases the major contribution is due to the
inertial drag of the boat. Similarly, boat length
must be carefully considered to avoid resonant
losses to wave drag. Coincidently, it has been
shown that a boat at race-pace minimises the
wave drag. However, further research into
changing wave drag conditions and also the
effects of aerodynamic drag would be useful.
Headwinds and crosswinds cause complex
turbulence effects which may have a
significant effect upon the boat speed.
The differential balance
characteristics between static and moving
boats have been discussed. Although a static
VIII was shown to be unstable, a moving boat
appears to be dynamically stable. This was
explained by considering the local exchange of
momentum with water as it flows either side of
the bows.
It was thought that two mechanisms
are responsible for the propulsion of the boat
during the stroke. In the earlier and later stages
of the stroke, when the angle between the oar
and the boat is small, the blade acts as a
hydrofoil. A leading-edge vortex is discharged
causing the creation bound vortex around the
blade which provides lift in a horizontal
direction, moving the blade forwards. In
contrast, as the blade reaches the middle of the
stroke it stalls in the water. In this phase, the
movement of a small amount of water at high
velocity produces acceleration. It has been
shown that the dynamical pressure generated
by the lift mechanism is nine times larger than
that generated from simply moving the water
with the blade during the middle of the stroke.
However, momentum transfer to the boat is
 more efficient during the middle of the stroke
because the longitudinal component of the
pressure is large. In contrast, for small blade
angles it is quite small. An exact comparison is
not appropriate until further research into the
longitudinal component of the lift force is
available. However, the mechanisms described
here explain why it is advantageous for the
rower to perform a quick catch and a quick
finish, as the time to shed a vortex (in the first
case), and the time the blade undergoes drag
(in the second case), is minimised.
In summary, the motion of rowing
boats has been discussed. The methods of
propulsion, while derived from the motion of
the rower, are variable. Furthermore, the drag
forces that oppose the motion of rowing boats
are complex, and their applicability often
depends upon the inherent conditions. Finally,
rowing boats have been shown to balance
stably when moving. Further research into
these phenomena is required to add to our
understanding of the motion of rowing boats;
in particular, the effects of aerodynamic drag,
lift forces upon acceleration, and 3D effects
such as twisting of the oar blade.

1. Dudhia, Basic Physics of Rowing, http://www.atm.ox.ac.uk/rowing/physics/ (2001)

2. L. Lazauskas, A Performance Prediction Model of Rowing Races,
http://www.cyberaid.net/library/rowing/stroke.htm (1997)
3. M. van Holst, Simulation of Rowing, http://home.hccnet.nl/m.holst/report.html (2004)
4. E.O. Tuck and L. Lazauskas, Low Drag Rowing Shells,
http://www.cyberaid.net/library/rowing/misbond/misbond.htm (1996)
5. S. Kerr, Balance of Racing Rowing Boats, http://www.btinternet.com/~furnivall.sc/fscbrb.htm (1998)
6. B. Carpenter, B. Krum, J. Thrasher, M. Epley, S. Nguyen, T. Pope, T. Blankenship, S. Sutherland, M.
Klara and P. Tidewll, Rowing Rigger Design,
http://filebox.vt.edu/eng/mech/tidwell/me4016/final/final.html (1997)
7. K. Young, Hydrodynamic Lift in the Sculling Stroke,
http://www.phys.washington.edu/~wilkes/post/temp/phys208/scull.lift.html (1997)
8. M. Brearley and N. J. de Mestre, Modelling the Rowing Stroke and Increasing its Efficiency, 3rd Conf.
Mathematics and Computers in Sport, Bond University, Queensland, Australia (1996)
9. J. Local, Discussion, Cambridge University Boat Club (2004)
10. M. Warner, Lectures on Fluids, http://www.tcm.phy.cam.ac.uk/~mw141/pt2/fluids.html (2004)

"When one rows it is not the rowing which moves the ship:
Rowing is only a magical ceremony by means of which one
compels a demon to move the ship".
F. Nietzsche
1b Rowing and materials science1

The introduction of a honeycomb sandwich structure in racing boats raised high

expectations in the racing community which, due to some teething problems, remained
partly unfulfilled. For some time the leading plastic shell manufacturers have been
striving to make use of the most advanced laminating method - heat cure - with little
success.

Our company was the first to offer you the full range of racing shells produced by the
process of heat cure lamination. Over the past sixteen years we have successfully
produced nearly 3500 boats from Single Sculls to Eights. These boats are enjoyed by
club oarsmen, oarswomen, scullers and top internationals alike.

The honeycomb sandwich, utilizing prepregnated

carbon fiber and glass cloths, is heated under vacuum
pressure and 'cooked' at 130 °C (290 °F). Due to the
heat and pressure, the materials fuse and the resulting
bond is 6-8 times stronger than the cold lamination.
The laminate is stable at all temperatures up to 90 °C
(190 °F), thus eliminating the major problems of
epoxy sandwich cold moulding - twisting and
blistering. The substantially increased strength of
bond, together with the use of carbon fiber, make the
hull more durable and far less susceptible to damage.
Finally the low, all up, weight of boats so eagerly
expected from honeycomb sandwich structure can be
achieved without adverse effect on the life of the
boat. For example, the weight of our 14 st (89 kg)
Sculling boat is 31 lb (141/2 kg), Double Sculls or
Coxless Pairs 58-62 lbs (26 /2 - 281/2 kg), etc. All of our boats are equipped with
1

adjustable fittings, rowing shoes and fully adjustable fittings, rowing shoes and fully
adjustable riggers to customer specifications.

The hulls are basically maintenance free and after sixteen years we have found no
deterioration of the structure whatsoever. Minor scratches are easily and quickly repaired
at low cost.

[1] www.janousek.co.uk
 2) -"There is only one way to row". 2a Technique
Volume 6 No 60 Rowing Biomechanics Newsletter March 2006

Ideas 9 We found that very often the sequence and ve-

locities of the segments on recovery mirrors the
9 The most popular classification of rowing sequence on the drive. If we plot the segments ve-
styles was introduced by Klavora in 1977 (1) and locities relative oar angle, they will look like mir-
defined three rowing styles: the Adam style; the ror images, where the negative part (recovery) re-
DDR style; the Rosenberg style: sembles positive part (drive). Below are charts of
ƒ Adam - Comparatively long legs drive and two rowers plotted relative oar angle:
limited amplitude of the trunk. Simultaneous Velocity (m/s) 3.0 Legs
activity of legs and trunk during the stroke; Trunk
2.0
ƒ DDR - Large, forward declination of the trunk, Arms
which begins the drive, followed by simultane- 1.0 Handle
ous activity of the legs; 0.0
ƒ Rosenberg - Large, forward declination of the -60 -40 -20
-1.0
0 20 40

trunk at the beginning of the stroke, then strong

leg extension without significant trunk activa- -2.0

tion. At the end of the cycle the trunk stops in -3.0 Angle (deg) 1
the deep backward position. Velocity (m/s) 3.0 Legs
We defined two main factors, which distin- Trunk
2.0
guish these styles: timing (simultaneous or conse- Arms
quent activity of two biggest body segments) and 1.0 Handle
emphasis during the drive (on legs or trunk). Then 0.0
we put these factors as X and Y axes of a quadrant: -60 -40 -20 0 20 40
-1.0
Trunk Emphasis -2.0

-3.0 Angle (deg)

DDR style Rosenberg style
Simultaneous Timing Consequent Timing
Grinko style
Adam style
Legs Emphasis
We found that the three styles perfectly fit
three quarters. However, we found that the fourth
rowing style must exist. This style has consequent
timing and emphasis on the legs drive. We called it
“Grinko style” after the name of talented Russian
coach Igor Grinko, who practises this style. Igor
coached many World champion scullers in USSR
and USA. One of them is Silver Olympic medalist
in M1x Jueri Jaanson (Appendix 1).
It is not very often we can see a pure example
of these rowing styles. Most of the rowers have a
style somewhere in between of these four extremi-
©2006 Dr. Valery Kleshnev, EIS, Bisham Abbey
ties.
2
The first rower prepares his trunk earlier dur-
ing recovery and approaches the catch with legs
only. The trunks is ready for the drive (trunk speed
is nearly zero). This rower has fast legs drive strait
after the catch and increase trunk velocity in the
second quarter of the drive. As we discussed in
RBN 2001/07 this “consequent“ rowing style pro-
duce higher relative maximal force and power.
The second rower spreads the trunks move-
ment across the recovery and continues tilting the
body until the last moment before the catch. As a
result, this rower “opens the body” early during the
drive and spreads its movement across the drive.
This “simultaneous” rowing style produces lower
maximal force and power, but the shape of force
curve is more rectangular.
An interesting practical application of this
principle could be the following: I you want to
achieve certain sequence and velocities of the seg-
ments during the drive, you should practice the
mirror sequence and velocities during recovery.
References
1. Klavora P. 1977. Three predominant styles: the Adam style; the
DDR style; the Rosenberg style. Catch (Ottawa), 9, 13.
Contact Us:
www.biorow.com e-mail: kleval@btinternet.com
 2b Race strategy / tactics

No.7 Volume 3 Rowing Biomechanics Newsletter July 2003

the most popular in silver (27.1%) and bronze
Back to basics (23.6%) medalists. In other words, if a crew put all
The RACE STRATEGY is defined as the efforts in the first 500m of the race, then the tactic
total distribution of crew effort during a race. It can would be “win or die”. If a crew saves energy for
be expressed as a sequence of four numbers the last 500m, then they have more chances to win
representing the ratio (%) of boat speed during each a medal, but fewer chances to win a gold medal.
500m section to the average boat speed over 2000m …this finding was confirmed by the analysis
for the crew. See RBN 10/2001, 10/2002 and (1) for in the pairs of competitors. In 61 cases (43.6%) the
more details on the race strategy. winners took the maximal advantage over the silver
RACE TACTICS are defined as a medalists during the first 500m section of the race:
distribution of crew efforts relative to other Gold w as w on at 1st 2nd 3rd 4th 500m
competitors in the race, and can be determined using
18.6%
two methods: 43.6%
• Relative to the average speed of all 22.9%
15.0%
competitors in the race, where ratios of individual
boat speed to the average of the race are produced for
each section; Silver w as w on at 1st 2nd 3rd 4th 500m
• Relative to the closest competitor. Five pairs
of place-takers were defined (1st-2nd, 2nd-3rd, …5th-6th) 32.9% 25.0%
and ratios of their boat speed were produced for each 22.1% 20.0%
section of the race.
In both methods, sequential numbers of the fastest
and slowest section relative to other competitors were Bronze w as w on at 1st 2nd 3rd 4th 500m
defined. Twelve possible combinations were
31.4% 25.7%
composed, called “matrix of race tactics” (1). For
example, tactic “1-4” means the first 500m section of 22.9% 20.0%
the race was the most successful and the last section
was the slowest relative to other competitors.
We analyzed race tactics of 14 Olympic boat In contrast, nearly one third of silver and bronze
types during the last 10 years. Some results are below. medalists had beaten their competitors at the final
500m section of the race.
Facts. Did you know that... …the majority of German (33.6%), British
…the most popular race tactics found were 4- (30.2%) and Romanian (31.7%) crews emphasized
1 (135 of 837 cases, 16.1%) and 1-4 (14.6%). the first section of the race. 38.6% of Australians,
Place Total 38.7% of Americans and 59.2% of French crews put
st nd rd th th th
Tactics 1 2 3 4 5 6 all efforts into the final section. Italians (32.2%) and
1-2 4 6 10 17 8 4 49 Canadians (30.4%) emphasized the second section.
1-3 8 9 12 12 14 7 62 This correlates with the percentage of gold medals
1-4 24 8 7 9 27 47 122 won by these countries (RBN 8/2001).
2-1 4 9 14 5 11 9 52
2-3 4 6 4 4 3 9 30 References
2-4 14 8 6 10 20 28 86 1. Kleshnev V. 2001. Racing strategy in
3-1 11 16 10 13 12 7 69 Rowing during Sydney Olympics. Australian Rowing.
3-2 11 3 5 4 8 2 33 24(1), 20-23.
3-4 19 9 6 7 9 3 53
4-1 20 38 33 26 13 5 135 Contact Us:
4-2 15 21 17 20 8 10 91 ©2002 Dr. Valery Kleshnev, AIS/Biomechanics
4-3 6 7 16 13 7 6 55 POBox 176, Belconnen, ACT, 2616, Australia
It is interesting that the tactic 1-4 was the most tel. (+61 2) 6214 1659, (m) 0413 223 290, fax: 6214 1593
popular in 1st (24 of 140 cases, 17.1%), 5th (19.3%) e-mail: kleshnevv@ausport.gov.au
and 6th (34.3%) place. In contrast, the 4-1 tactic was
Further information on:
• Propulsion
• Water resistance – viscosity
• Kinetics
• Center of mass – balance
• Levers
…and more
http://www.atm.ox.ac.uk/rowing/physics

• Simulation of hydrodynamic forces on the blades and more

http://home.hccnet.nl/m.holst/RoeiWeb.html

Also:
An insider's view…
http://www.youtube.com/watch?v=B2EJ5CVb7pk

A spectacular finish. Notice the commentator, especially near the end of the
race. (No, it is not Manolis Mavrommatis).
http://www.youtube.com/watch?v=D6rHgCUWIE4

“Real athletes row. Everyone else just plays games”.